Is this even possible? Perception of direction without any perception of distance? That case would probably lead to localization inside the head.
Again you are right, there can be no localization without estimation of
distance, i fear i just wanted to neglect how distance was estimated in
my case of listening to the anechoic recording.
Markus: You force me to think after writing, what i should have done
before instead. How can you do that to me ?
I am usually thinking of Initial Time Delay Gap as being a cue regarding
distance of a natural sound source (like a kettle drum) in a room
(like a concert hall).
ITDG Animation: Anfangszeitlücke ITDG
While ITDG is the time difference in arrival between direct sound and
the first strong reflection, ITDG may be interpreted as being infinitely
long in case reflections are missing completely.
Which in turn could be perceived as the sound source being coincident
with the listener or at least the sound source and the listener b o t h
being outside of any room and away from any reflecting surfaces
(like hanging in the air freely).
Nevertheless, since most sound sources have a frequency dependent radiation
pattern, the distance of a (natural) sound source (type) known to listener
may also be estimated by change of (expected) tonality (spectrum of direct
sound changing with distance), even if no IDTG cues are available. That
mechanism would still be in effect without ITDG cues (e.g. if no or only
too weak/too diffuse reflections are present).
So if in my case no in head localization took place, i have to ask where
the cues for the phantom sources having any distance came from, right ?
So i first have to describe how the impression of distance of the phantom
sources really was:
- pretty close but in a strange way also "undefined".
- i could shift them in distance mentally, which is not possible with
usual recordings in the same way.
- but never a phantom source could be imagined as being "far away" like
the typical orchestral kettle drum sound at most real venues and also
on most recordings.
- there was no d i f f e r e n c e perceivable between instruments, in that
some instruments would be closer than others.
At least there was the impression, there is "some" distance and also "some room".
But the distance felt "short" and the room very very dry ... far too dry for
being my own listening room, if you know what i mean.
That was astonishing to me, it seemed that the virtual room in which the
instruments (phantom sources) where placed was perceived as being far more
dry than my listening room, which i know quite well from talking or handclapping.
I can only imagine 3 kinds of cues making up any distance perceived:
- the recording venue not being "quite anechoic"
- my own listening room delivering "substitutional ITDG cues" which were
processed instead of those expected on a typical recording
- the microphone(s) during recording captured the sound of each instrument at a
certain distance and the tonality of the recorded sound itself being the cue
________________________________________________
Situation in my current listening room:
The early ipsilateral reflections from the side walls should be rather weak,
since there is the dipole notch of the speakers used.
The first strong reflections from the front wall should arrive at about 10ms,
as i am using dipoles having about 1,8m distance to the front wall.
Nevertheless i am diffusing those front reflections into
- direct arrival (1st order frontal) and
- "over the corner arrivals" ( 2nd order ipsilateral and 3rd order contralateral)
which is done by placement.
- additionally a diffuser/reflector is placed in the corner directing some energy
to the opposite sides causing e.g. 2nd order contralateral reflections.
- there is also an absorber cushion at each lower corners behind the speakers
The first strong arrivals should be smeared quite well around 10 ... 20ms.
________________________________________________
In fact the ITDG of the listening room has to be larger than that of the recording
venue, for the IDTG of the recording venue to be perceivable during loudspeaker
reproduction i.e. the recording venue being r e p r o d u c i b l e in terms of
typical phantom source distances.
IMO that means, some early reflections in the listening room may be allowed, but
they should in no way carry much energy and also should be sufficiently diffuse.
The strongest arrivals (if not sufficiently diffused) should have an ITDG in
the order of known concert halls or (better) longer.
ITDG in a concert hall is of course dependent on source and listener (microphone)
position, it is not a room constant. But averages can be found in the literature
to even compare concert halls.
One major difference between the ITDG in a concert hall and that in a listening
room - especially when using coherent loudspeakers over the whole audible
bandwith - is that the ITDG in a concert hall is usually widely spread among
different groups and even single instruments and voices.
Thus perceiving "offset in depth" is supported, while the ITDG cues of a common
speaker/room system are by far more uniform/nondiffuse and will tend to
flatten any "offset in depth" captured on a good recording.
Since fullrange coherent loudspeakers (especially conventional CD concepts like
e.g. proposed by SL above) tend to cause reflections strongly correlated to the
direct sound in usual (non extensively treated) listening rooms, those
"reproduction induced ITDG cues" are so very detrimental to imageing, because
they strongly c o m p e t e with ITDG cues on the recording.
Instruments and voices in a real musical venue never compete this way: They keep
separable by producing individual sets of ITDG cues and also patterns of later
reflections separable in arrival time and direction.
The conventional loudspeaker (fullrange coherent and also optionally CD) is a
bad musical instrument in that respect: It competes with all the instruments and
voices to be reproduced by causing strong and uniform ITDG cues combined even with
uniform later reflections strongly correlated with the direct sound over the whole
audible bandwidth.
Such loudspeakers force the user more than it is desirable to build a reflection
free zone around the speakers in order to have long ITDG in the listening room.
What is left from strong reflections arriving at the listener has to be treated
with diffusers and absorbers, which is not a practical way to go consequently
for most listeners.
Some ITDG data for known concert halls:
http://www.sengpielaudio.com/WasBedeutetAnfangszeitluecke.pdf
distance, i fear i just wanted to neglect how distance was estimated in
my case of listening to the anechoic recording.
Markus: You force me to think after writing, what i should have done
before instead. How can you do that to me ?
I am usually thinking of Initial Time Delay Gap as being a cue regarding
distance of a natural sound source (like a kettle drum) in a room
(like a concert hall).
ITDG Animation: Anfangszeitlücke ITDG
While ITDG is the time difference in arrival between direct sound and
the first strong reflection, ITDG may be interpreted as being infinitely
long in case reflections are missing completely.
Which in turn could be perceived as the sound source being coincident
with the listener or at least the sound source and the listener b o t h
being outside of any room and away from any reflecting surfaces
(like hanging in the air freely).
Nevertheless, since most sound sources have a frequency dependent radiation
pattern, the distance of a (natural) sound source (type) known to listener
may also be estimated by change of (expected) tonality (spectrum of direct
sound changing with distance), even if no IDTG cues are available. That
mechanism would still be in effect without ITDG cues (e.g. if no or only
too weak/too diffuse reflections are present).
So if in my case no in head localization took place, i have to ask where
the cues for the phantom sources having any distance came from, right ?
So i first have to describe how the impression of distance of the phantom
sources really was:
- pretty close but in a strange way also "undefined".
- i could shift them in distance mentally, which is not possible with
usual recordings in the same way.
- but never a phantom source could be imagined as being "far away" like
the typical orchestral kettle drum sound at most real venues and also
on most recordings.
- there was no d i f f e r e n c e perceivable between instruments, in that
some instruments would be closer than others.
At least there was the impression, there is "some" distance and also "some room".
But the distance felt "short" and the room very very dry ... far too dry for
being my own listening room, if you know what i mean.
That was astonishing to me, it seemed that the virtual room in which the
instruments (phantom sources) where placed was perceived as being far more
dry than my listening room, which i know quite well from talking or handclapping.
I can only imagine 3 kinds of cues making up any distance perceived:
- the recording venue not being "quite anechoic"
- my own listening room delivering "substitutional ITDG cues" which were
processed instead of those expected on a typical recording
- the microphone(s) during recording captured the sound of each instrument at a
certain distance and the tonality of the recorded sound itself being the cue
________________________________________________
Situation in my current listening room:
The early ipsilateral reflections from the side walls should be rather weak,
since there is the dipole notch of the speakers used.
The first strong reflections from the front wall should arrive at about 10ms,
as i am using dipoles having about 1,8m distance to the front wall.
Nevertheless i am diffusing those front reflections into
- direct arrival (1st order frontal) and
- "over the corner arrivals" ( 2nd order ipsilateral and 3rd order contralateral)
which is done by placement.
- additionally a diffuser/reflector is placed in the corner directing some energy
to the opposite sides causing e.g. 2nd order contralateral reflections.
- there is also an absorber cushion at each lower corners behind the speakers
The first strong arrivals should be smeared quite well around 10 ... 20ms.
________________________________________________
In fact the ITDG of the listening room has to be larger than that of the recording
venue, for the IDTG of the recording venue to be perceivable during loudspeaker
reproduction i.e. the recording venue being r e p r o d u c i b l e in terms of
typical phantom source distances.
IMO that means, some early reflections in the listening room may be allowed, but
they should in no way carry much energy and also should be sufficiently diffuse.
The strongest arrivals (if not sufficiently diffused) should have an ITDG in
the order of known concert halls or (better) longer.
ITDG in a concert hall is of course dependent on source and listener (microphone)
position, it is not a room constant. But averages can be found in the literature
to even compare concert halls.
One major difference between the ITDG in a concert hall and that in a listening
room - especially when using coherent loudspeakers over the whole audible
bandwith - is that the ITDG in a concert hall is usually widely spread among
different groups and even single instruments and voices.
Thus perceiving "offset in depth" is supported, while the ITDG cues of a common
speaker/room system are by far more uniform/nondiffuse and will tend to
flatten any "offset in depth" captured on a good recording.
Since fullrange coherent loudspeakers (especially conventional CD concepts like
e.g. proposed by SL above) tend to cause reflections strongly correlated to the
direct sound in usual (non extensively treated) listening rooms, those
"reproduction induced ITDG cues" are so very detrimental to imageing, because
they strongly c o m p e t e with ITDG cues on the recording.
Instruments and voices in a real musical venue never compete this way: They keep
separable by producing individual sets of ITDG cues and also patterns of later
reflections separable in arrival time and direction.
The conventional loudspeaker (fullrange coherent and also optionally CD) is a
bad musical instrument in that respect: It competes with all the instruments and
voices to be reproduced by causing strong and uniform ITDG cues combined even with
uniform later reflections strongly correlated with the direct sound over the whole
audible bandwidth.
Such loudspeakers force the user more than it is desirable to build a reflection
free zone around the speakers in order to have long ITDG in the listening room.
What is left from strong reflections arriving at the listener has to be treated
with diffusers and absorbers, which is not a practical way to go consequently
for most listeners.
Some ITDG data for known concert halls:
http://www.sengpielaudio.com/WasBedeutetAnfangszeitluecke.pdf
Last edited:
it's rather a kind of "proof" that Moulton is right when He says that any typical listening room becomes just a carrier for the recorded information and that professor Hartmann is right with His hypothesis that "recorded reverberation is far more audible than live reverberation"
the floor is the answer
in every listening room in every conventional stereo setup there is the floor reflection and it is audible and it makes difference
in fact the above is just all this thread and FCUFS is about because it's the floor reflection that determines the ITDG in every listening room in every conventional stereo setup
Last edited:
Hi graaf,
in my special case, this is the speaker's step response
http://www.dipol-audio.de/model2-messungen-dateien/model2_nearfield_step.JPG
The speaker is a floorstander with semi rigid bending wave membrane.
The graph is including floor reflection: There is virtually none.
Membrane is about 40x110 cm and nearly touches the floor.
There are multiple excitation points and bending waves on the
membrane propagate faster than sound in air from upper midrange
on upwards.
No issues concerning floor reflections and - to be honest - it was
one of the main goals in developing this speaker, to circumvent floor
reflection as far as possible.
in my special case, this is the speaker's step response
http://www.dipol-audio.de/model2-messungen-dateien/model2_nearfield_step.JPG
The speaker is a floorstander with semi rigid bending wave membrane.
The graph is including floor reflection: There is virtually none.
Membrane is about 40x110 cm and nearly touches the floor.
There are multiple excitation points and bending waves on the
membrane propagate faster than sound in air from upper midrange
on upwards.
No issues concerning floor reflections and - to be honest - it was
one of the main goals in developing this speaker, to circumvent floor
reflection as far as possible.
Last edited:
Three measurements of an early "Model2" version from 2012
placed in the listening room mentioned.
Of course FR changed noteably up to now but gross behaviour
due to polar pattern or decay has not changed.
1) Setting:
Microphone on the bottom in front of speaker.
Mic lying on a large cushion additionally.
Floor reflection supressed acoustically this way.
Sidewall reflection damped by another cushion.
Frontwall reflection gated.
2) Setting:
Microphone on 70cm height over naked floor (wood tiles).
Floor reflection not supressed.
Sidewall reflection not supressed.
Frontwall reflection gated.
3) Setting:
Microphone on 70cm height over naked floor (wood tiles).
Floor reflection not supressed.
Sidewall reflection not supressed.
Frontwall reflection not supressed.
Later Reflections gated.
____________________________
Please note when going from 1) to 2) the axis is shifting and FR changes.
The effect of floor and also sidewall reflections is rather unoffensive with
that speaker. Note: Floor and sidewalls of the room are untreated in proximity
to the speaker, just the frontwall and front corners have some diffusers.
placed in the listening room mentioned.
Of course FR changed noteably up to now but gross behaviour
due to polar pattern or decay has not changed.
1) Setting:
Microphone on the bottom in front of speaker.
Mic lying on a large cushion additionally.
Floor reflection supressed acoustically this way.
Sidewall reflection damped by another cushion.
Frontwall reflection gated.
2) Setting:
Microphone on 70cm height over naked floor (wood tiles).
Floor reflection not supressed.
Sidewall reflection not supressed.
Frontwall reflection gated.
3) Setting:
Microphone on 70cm height over naked floor (wood tiles).
Floor reflection not supressed.
Sidewall reflection not supressed.
Frontwall reflection not supressed.
Later Reflections gated.
____________________________
Please note when going from 1) to 2) the axis is shifting and FR changes.
The effect of floor and also sidewall reflections is rather unoffensive with
that speaker. Note: Floor and sidewalls of the room are untreated in proximity
to the speaker, just the frontwall and front corners have some diffusers.
Attachments
Last edited:
Seems we've struck an accord of sorts...
1. A single driver covering the critical range, with excellent off axis response. I like to stick with 200-2k in a mid personally. 2k seems to be critical.
2. Use of two such drivers side by side (truncated) frame edge to (truncated) frame edge producing interaural compensation (and comb filtering... WHICH I HEARD NONE OF faced upward! Oh the shock, horror, will it never end! LOL
3. Tweeter crossed @2.2k
Found that the tweeter placement did not need to be faced upward and did not take away from this effect in the slightest even when on axis at typical heigth. It was time aligned with the acoustic center @2.2k
LA, I've been hard pressed as to what the balance is, perhaps 4.5dB (or more) greater off axis than on in the critical range. Might be a good starting point.
Indeed more measurements are necessary. We are finding too many commonalities to be meer coincidence
1. A single driver covering the critical range, with excellent off axis response. I like to stick with 200-2k in a mid personally. 2k seems to be critical.
2. Use of two such drivers side by side (truncated) frame edge to (truncated) frame edge producing interaural compensation (and comb filtering
... WHICH I HEARD NONE OF faced upward! Oh the shock, horror, will it never end! LOL3. Tweeter crossed @2.2k
Found that the tweeter placement did not need to be faced upward and did not take away from this effect in the slightest even when on axis at typical heigth. It was time aligned with the acoustic center @2.2k
LA, I've been hard pressed as to what the balance is, perhaps 4.5dB (or more) greater off axis than on in the critical range. Might be a good starting point.
Indeed more measurements are necessary. We are finding too many commonalities to be meer coincidence
A funny device:
Voigt Domestic Corner Horn Reflectors X2 for Lowther Fostex Loudspeaker Drivers | eBay
Would need some Basotect to prevent wall reflections.
Voigt Domestic Corner Horn Reflectors X2 for Lowther Fostex Loudspeaker Drivers | eBay
Would need some Basotect to prevent wall reflections.
still externalization can be caused by lateral reflections or just by HRTF from the speakers, in such a case there is no distance cue though hence this paradoxical sense of a sound source being there outside yet at no particular, defined distance
http://www.hovirinta.fi/audio/psykoakustiikka/aiheet/direct6.gif
File:Akustik - Richtungsbänder.svg - Wikimedia Commons
The Directional Bands found by Jens Blauert, are correlated with
the Head Related Transfer Function. Especially the band for
"above" direction correlates with a peak in most person's pinna
transfer function around 7-8Khz, which occurs if sound comes
from "above" direction.
Directional Bands mean the spectral information is used to estimate
the direction of a sound source in the median plane, where no
interaural differences can be evaluated by our brain.
There are characteristic directional bands for
"front", "rear", "above" directon.
The other way round increased spectral energy within a certain
directional band (say "above"), may confirm the impression of a
sound source being elevated by some angle (which need not be the case in reality).
____________
I experimented a lot with practical use of Directional Bands too.
Since i read from Directional Bands more than 2 decades ago in
Jens Blauert's book on "Spatial Hearing" i felt that directional
bands are no good crossover frequencies as such.
That is especially for multiway speakers without matched DI at
crossover frequency. You usually cannot avoid some anomalies
at crossover like e.g. in
- DI or
- shift in main axis of radiation or
- power response
It is my opinion, that it is best to have crossover frequencies
in the "neutral" ranges between the directional bands.
Pick your preferred 2,2Khz and tell me, what you find in the graphs.
Me thinks too, that 2,1 ... 2,3 Khz is a good range for crossing
over in a 2-way.
This is about how to minimize "colorations" or "errors" perceived
in physiological units, while the "objective" error (e.g. peak/dip
in power response measured in decibels and bandwidth) may stay the
same if you chose a different frequency.
To me the best way is to design a speaker according to some
physiological "rules" (if available) but also optimize technically:
The drivers should be optimized for that crossover frequency and IMO
abrupt changes in DI over frequencies should be avoided:
If you get a multiway that perfect, that you cannot find the crossover
point by measuring acoustic bahaviour from outside, you may also
put the crossover point in the mid of a Directional Band ... it is
just that i have never seen such a multiway speaker.
And this is why i try to find alternatives to usual multiway seriously
since about 2006. The common 4" to 10" "fullrangers" are not an
alternative in itself to me.
And btw. "upfiring highs" is not necessary IMO too to circumvent
artefacts arising from floor reflections.
Last edited:
Greebster, i did not quite understand that. Could you please point out
which kind of balance you mean and
where the 4.5db difference in level should occur ?
That is exactly describing, what my auditive impression was.
And i can live with the explanation very well as a hypothesis.
Oliver,
the ITDG is no good measure for distance cues in a room. According to Bronkhorst "Modeling auditory distance perception in rooms" our depth perception is (to a large part) a function of the relation between reverberant energy and direct energy in the room. Human distance perception by ear is notoriously bad, still it becomes better when more reflecting walls are included.
In recent discussions we have found a good analogy for stereo reproduction in the metapher of "listening through a window into the original recording." This should apply well enough to recordings in a music hall environment. When listening, we receive two independent sets of direct/reverberant energy information: One from the listening room and one from the recording.
The set from the listening room places the "window" (as mentioned above) at its appropriate distance in our room. The set from the recording places the instruments/voices and hall dimensions behind that "window".
We must make a clear difference between both sets. They don't mix, if we listen to a large hall recording in our small place. They will mix, if we listen to an intimate near field recording of a small ensemble. That's because the direct/reverberant informations are vastly different in the first case and become comparable in the second case. This is one aspect of the "we are there" against "they are here" controversity.
The other aspect is the placement of the "window" in our room:
Stereo speakers with very high directivity (high direct/reverberant ratio) will place the window directly at our head. The listening room influence is highly subdued and we may have the front soloist playing/singing at nose distance. This enhances the impression that all recorded venues at least "include" the listening room.
Stereo speakers with conventional directivity might place that window at the loudspeaker basis.
Speakers with a certain degree of rearward energy (dipoles) will place the window somewhere behind the loudspeaker basis.
For discussions sake I would like to introduce the "Transparency" of this window. If single reflections in the listening room become very prominent/dominant, they make the window "less transparent" - reducing the depth of the auditory scene behind it. In extreme cases (easily happening with fully symmetrical dipoles and no treatment of a totally reflective front wall) the "window" turns into a "projection screen" with no apparent depth left.
This has two consequencies: Prevent early reflections from sticking out (level wise) from the rest, if you want to keep the window "clear". And reduce the overall D/R ratio, if you want the place the "window" close to you.
the ITDG is no good measure for distance cues in a room. According to Bronkhorst "Modeling auditory distance perception in rooms" our depth perception is (to a large part) a function of the relation between reverberant energy and direct energy in the room. Human distance perception by ear is notoriously bad, still it becomes better when more reflecting walls are included.
In recent discussions we have found a good analogy for stereo reproduction in the metapher of "listening through a window into the original recording." This should apply well enough to recordings in a music hall environment. When listening, we receive two independent sets of direct/reverberant energy information: One from the listening room and one from the recording.
The set from the listening room places the "window" (as mentioned above) at its appropriate distance in our room. The set from the recording places the instruments/voices and hall dimensions behind that "window".
We must make a clear difference between both sets. They don't mix, if we listen to a large hall recording in our small place. They will mix, if we listen to an intimate near field recording of a small ensemble. That's because the direct/reverberant informations are vastly different in the first case and become comparable in the second case. This is one aspect of the "we are there" against "they are here" controversity.
The other aspect is the placement of the "window" in our room:
Stereo speakers with very high directivity (high direct/reverberant ratio) will place the window directly at our head. The listening room influence is highly subdued and we may have the front soloist playing/singing at nose distance. This enhances the impression that all recorded venues at least "include" the listening room.
Stereo speakers with conventional directivity might place that window at the loudspeaker basis.
Speakers with a certain degree of rearward energy (dipoles) will place the window somewhere behind the loudspeaker basis.
For discussions sake I would like to introduce the "Transparency" of this window. If single reflections in the listening room become very prominent/dominant, they make the window "less transparent" - reducing the depth of the auditory scene behind it. In extreme cases (easily happening with fully symmetrical dipoles and no treatment of a totally reflective front wall) the "window" turns into a "projection screen" with no apparent depth left.
This has two consequencies: Prevent early reflections from sticking out (level wise) from the rest, if you want to keep the window "clear". And reduce the overall D/R ratio, if you want the place the "window" close to you.
Wrong assumption. The original Karlson with good coaxial and appropriate crossover is as good or better than some other speakers I've owned including: Edgarhorn System 100, full size Acoustat with built in tube amp, Magnaplanar Tympani 1D, Klipschorn, etc. The Karlson design has good dispersion, reduction of cone excursion, and reduction of modulation distortion.
Rudolf
I have not visited this thread because I did not find it having much to say. But I think that your explanation above is quite excellent. This is very much how I see the whole thing as well including the discussion of the "window" being dirty or what not. The window could also be stained glass or frosted and make the scene appear more "pleasing", more "diffuse", but that is clearly not "accuracy". The "mixing" of room and recording acoustics is also right on the money IMO.
Good job!
Freddi - you are like a broken record- I remember these same Karlson claims for several other forums. At least you are diligent if not believable.
I have not visited this thread because I did not find it having much to say. But I think that your explanation above is quite excellent. This is very much how I see the whole thing as well including the discussion of the "window" being dirty or what not. The window could also be stained glass or frosted and make the scene appear more "pleasing", more "diffuse", but that is clearly not "accuracy". The "mixing" of room and recording acoustics is also right on the money IMO.
Good job!
Freddi - you are like a broken record- I remember these same Karlson claims for several other forums. At least you are diligent if not believable.
And when placed on the floor, a Karlson still sounds like it is on the floor, sharing the same problem I found with any FCUFS.
That said, for those that like all the extra reflected sound a FCUFS offers, the Karlson adds even more extra reflections from the distinctively shaped (Christmas tree) band-pass opening.
After later listening as an adult I far preferred the Karlsons I grew up with (that Freddie now owns) to our other household stereo, a Magnavox console with forward facing woofers and side firing Heppner horns.
That Magnavox certainly did sound "spacious" if you laid down on your back with head between the woofers, to this day I remember never hearing Jimi Hendrix the same way again as that moment around 1969 when I tried that.
Art
Last edited:
- well, I tend to back up things more than some stuff promulgated in forums. IMO there's more BS speaker activity happening than just my old friend the Karlson. what's the significance of distortion measurements in general when music is considered to consist of transients?
Eminence Beta15cx swapped between a Karlson and a reflex the size of the Karlson's rear chamber , each with similar fb (~48-50 ballpark)
An externally hosted image should be here but it was not working when we last tested it.
did I misunderstand this thread ? - why does the Karlson not fit the thread description?
Last edited: