I like that one!
The disappearing speaker is a great trick, I really like it. Mine mostly do it.
Two speakers and a head form basis of mechanical system. Complete mechanical system involves room. All cues originate from speakers; garbled cues come from speakers and room. All else is perceptual illusion.
Bang for the buck remains two channel playback.
What I described is completely separable from room reflections.
An externally hosted image should be here but it was not working when we last tested it.
Maximal misalignment is 1/3 wave at 1kHz crossover point along vertical axis. Woofer is completely omnidirectional in horizontal plane. Tweeter remains fairly omnidirectional to about 2.5kHz. Speaker associated reflections are minimal. Imaging remains spectacular even at 18".
Performance links:
Active v Passive #347
How to achieve coherence #43
How to achieve coherence #48
Regards,
Andrew
If you like to explain it that way, here is my version: I have no problem listening "clearly" through the lens (tweeter). But I can identify the lens (tweeter) between me and the music, if I want so. Takes me some concentration though.
Rudolf
Voice coils are roughly on vertical access. Delay takes care of rest of offset.
If you mean coaxial to woofer on listening axis, the simple answer is ease of application.
Regards
Andrew
Hi Art, all
You are exactly right Art.
Good imaging speakers produce a strong phantom center image if driven with a mono signal while speakers that radiate clues about where they are, produce a mono phantom image but also a Right and left source identity.
In the later case, the sound with stereo is a panorama but not an image of “a” source or sources in space. When one can produce a mono phantom image without the side images, then one can also produce an apparent source location anywhere in between.
A greatly unappreciated factor is that how we hear a single image in our head isn’t very much like like how a microphone detects pressure.
Our ears have a response that is nothing like flat (google equal loudness curves) and there are major changes that occur depending on the angle of the incoming sound. Our brain / ear / lifetime of hearing experience, converts the two ear signals into one acoustic image. All the goofiness in our ears behavior like our Pina response etc is how we can tell what direction a source of sound is.
The problem is, a single microphone does not capture that one image, it only captures the pressure where the mic element is.
Thus, a very important aspect of what makes an image is not captured with even the best measurement microphone.
In the development of the Synergy horn (linesource has an SH-25 in his picture) I was working on the SH-50 at home in my listening room. I should back up slightly. I was working and still am for a company that makes loudspeakers for commercial sound, large scale installations etc.
Here, one of the biggest enemies is a lack of pattern control, you want the sound to be projected where the ears are and NOT the walls or ceiling.
A basic problem is when you move the speakers farther away from the ears, they have to be more powerful and to be more powerful, it takes more drivers and more frequency ranges.
The problem is, two or more source of sound can combine coherently into one new source like close together subwoofers do, but only if they are less than a quarter wavelength apart. At a half wavelength or greater spacing, the two sources now radiate independently and what you get depends on where you are relative to the two sources. Here, with two sources producing the same frequency (like in a speaker at at crossover or with multiple drivers) you find anywhere you have a difference in path lengths of 1 or more odd half wavelengths, you have a cancellation notch.
A polar pattern shows lobes and nulls, the hallmark of an acoustic interference pattern.
In the scale we work in, it is the huge amount of energy that the large speaker systems radiate outside the pattern is one reason large scale sound is so bad, say you have 80 source of source at varying distance from each other and in spite of being very large, it radiates a lot of energy to the rear, to the sides and behind because it generates a huge interference pattern. Anywhere in front of it, a single impulsive input signal arrives spread out in time with energy arriving from the closest sources first and farthest last, this also means the frequency response is different everywhere.
If you want to play with how sound sources interact, there are two free programs on our website that are fast and easy for that stuff. Also Doug Jones who was acoustics department chair at Columbia College, has made a couple “how to” videos. Both are free but try the 2D version first.
DDT Files | Danley Sound Labs, Inc.
Anyway It is the texture of that interference pattern that you hear when you hear a large concert array outdoors and the wind blows. It is an interference pattern that you hear when you place two full range loudspeakers side by aide and play music or better yet pink noise and walk past them.
I am pretty sure It is that texture, the difference between what arrives at your right and left ears from ONE speaker what gives a speaker it’s (what called) spatial identity.
I was working on a wide band constant directivity horn, this was driven by the realization that a conical horn had poor lf loading because it’s expansion rate (which governs the “high pass” effect of horn transformation tied to the rate of expansion) was fixed in an exponential horn a variable in a conical horn. At the apex, the expansion was very rapid and would only couple high frequencies BUT a short distance down the horn, the expansion was suitable for mid and further down low frequencies.
The answer I thought was to try mounting drivers suitable for mid range horn loading on the sides of a horn and see what they do. The rules eventually boiled down to a few simple and a few not so obvious. Where two or more drivers interact, the distance between them must be less than about ¼ wavelength at the highest frequency of interest and you can’t drive the horn from a dimension where it is more than about a quarter wavelength across at the highest frequency of interest.
Anyway I had found some rules for making multiple drivers couple coherently into one with no lobes and nulls and was working on a crossover which took advantage of the front to back driver position being more or less inverse of a normal crossover phase shift or its steps in group delay.
I am a fan if Dick Heysers work and was an early TEF user so while acoustically it was one acoustic source, I wanted to make it one source in time as well, eliminating the effect of a crossover on phase / group delay..
As I got the acoustic part and crossover better and better, I heard a weird thing. With one speaker on, playing a voice quietly and then closing my eyes, it was still easy to hear where the speaker was left right, up down BUT it got harder to hear its position in depth. When that happened, it sounded more like it was as far away as the cues in the recording made it sound. It reminded me of an experience when I was fixing my bosses esl-63’s where the voice sounded like it was behind the speaker.
The reason is at least in part because they both radiate a simple segment of a sphere, from the same location in time and space, over a wide bandwidth. When that happens, what arrives at the right and left ear from one speaker is “as if” it was from the same source. One can only localize the depth location of a speaker IF there is enough difference in the speakers radiation between what arrives at your right and left ears. If what arrives is similar enough, then we hear the clues in the recording more strongly.
In large scale, the lack of an interference pattern is VERY audible, particularly as the wind blows, hardly anything happens to the sound and the frequency response shape is essentially identical everywhere (is different everywhere with an interference pattern).
I have noticed too that small full range drivers on a large flat baffle produce a similar effect, over much of their range, they radiate as a simple spherical segment and not an interference pattern.
It is interesting that Barleywaters arrangement being so small it would also radiate as simple spherical pattern as well up to tweeter directivity.
While it is known speakers sound better when you deal with edge diffraction and such, I am not sure people realize that is also one way to produce an interference pattern, re-radiation where there is an abrupt acoustic impedance change. Horn are the ultimate extension of proving a controlled boundary out to a large dimension where a wave can freely propagate at the same angle (as defined by Don Keeles pattern loss thumb rule).
I believe what is boils down to at least spatially speaking, is that with only one speaker on, the more similar the signal reaching both ears, the less information is carried about the speaker as the source in physical space. When the speaker radiates little about itself, then it’s harder to hear distance to the speaker, the more strong the mono phantom image is / the less able you hear a right and left sources and when the source identity drops away, then you have more information arriving intact from the recording..
If you have headphones, here is video a customer of ours took of two pairs of speakers he owns, the first part shows the source identity thing.
Danley Sound Labs SM80 vs. d&b Q7 - YouTube
They have a very similar frequency response (one is eq to match the other ) so the spectral content / balance is very similar. When he switches back and forth, you can hear a pretty clear difference in how it sounds but it isn’t anything to do with frequency response but how the two sources radiate in time and space. What you hear has to do with the stereo image and the source identity although I am not sure I could explain it if you haven’t heard it.
Best,
Tom Danley
You are exactly right Art.
Good imaging speakers produce a strong phantom center image if driven with a mono signal while speakers that radiate clues about where they are, produce a mono phantom image but also a Right and left source identity.
In the later case, the sound with stereo is a panorama but not an image of “a” source or sources in space. When one can produce a mono phantom image without the side images, then one can also produce an apparent source location anywhere in between.
A greatly unappreciated factor is that how we hear a single image in our head isn’t very much like like how a microphone detects pressure.
Our ears have a response that is nothing like flat (google equal loudness curves) and there are major changes that occur depending on the angle of the incoming sound. Our brain / ear / lifetime of hearing experience, converts the two ear signals into one acoustic image. All the goofiness in our ears behavior like our Pina response etc is how we can tell what direction a source of sound is.
The problem is, a single microphone does not capture that one image, it only captures the pressure where the mic element is.
Thus, a very important aspect of what makes an image is not captured with even the best measurement microphone.
In the development of the Synergy horn (linesource has an SH-25 in his picture) I was working on the SH-50 at home in my listening room. I should back up slightly. I was working and still am for a company that makes loudspeakers for commercial sound, large scale installations etc.
Here, one of the biggest enemies is a lack of pattern control, you want the sound to be projected where the ears are and NOT the walls or ceiling.
A basic problem is when you move the speakers farther away from the ears, they have to be more powerful and to be more powerful, it takes more drivers and more frequency ranges.
The problem is, two or more source of sound can combine coherently into one new source like close together subwoofers do, but only if they are less than a quarter wavelength apart. At a half wavelength or greater spacing, the two sources now radiate independently and what you get depends on where you are relative to the two sources. Here, with two sources producing the same frequency (like in a speaker at at crossover or with multiple drivers) you find anywhere you have a difference in path lengths of 1 or more odd half wavelengths, you have a cancellation notch.
A polar pattern shows lobes and nulls, the hallmark of an acoustic interference pattern.
In the scale we work in, it is the huge amount of energy that the large speaker systems radiate outside the pattern is one reason large scale sound is so bad, say you have 80 source of source at varying distance from each other and in spite of being very large, it radiates a lot of energy to the rear, to the sides and behind because it generates a huge interference pattern. Anywhere in front of it, a single impulsive input signal arrives spread out in time with energy arriving from the closest sources first and farthest last, this also means the frequency response is different everywhere.
If you want to play with how sound sources interact, there are two free programs on our website that are fast and easy for that stuff. Also Doug Jones who was acoustics department chair at Columbia College, has made a couple “how to” videos. Both are free but try the 2D version first.
DDT Files | Danley Sound Labs, Inc.
Anyway It is the texture of that interference pattern that you hear when you hear a large concert array outdoors and the wind blows. It is an interference pattern that you hear when you place two full range loudspeakers side by aide and play music or better yet pink noise and walk past them.
I am pretty sure It is that texture, the difference between what arrives at your right and left ears from ONE speaker what gives a speaker it’s (what called) spatial identity.
I was working on a wide band constant directivity horn, this was driven by the realization that a conical horn had poor lf loading because it’s expansion rate (which governs the “high pass” effect of horn transformation tied to the rate of expansion) was fixed in an exponential horn a variable in a conical horn. At the apex, the expansion was very rapid and would only couple high frequencies BUT a short distance down the horn, the expansion was suitable for mid and further down low frequencies.
The answer I thought was to try mounting drivers suitable for mid range horn loading on the sides of a horn and see what they do. The rules eventually boiled down to a few simple and a few not so obvious. Where two or more drivers interact, the distance between them must be less than about ¼ wavelength at the highest frequency of interest and you can’t drive the horn from a dimension where it is more than about a quarter wavelength across at the highest frequency of interest.
Anyway I had found some rules for making multiple drivers couple coherently into one with no lobes and nulls and was working on a crossover which took advantage of the front to back driver position being more or less inverse of a normal crossover phase shift or its steps in group delay.
I am a fan if Dick Heysers work and was an early TEF user so while acoustically it was one acoustic source, I wanted to make it one source in time as well, eliminating the effect of a crossover on phase / group delay..
As I got the acoustic part and crossover better and better, I heard a weird thing. With one speaker on, playing a voice quietly and then closing my eyes, it was still easy to hear where the speaker was left right, up down BUT it got harder to hear its position in depth. When that happened, it sounded more like it was as far away as the cues in the recording made it sound. It reminded me of an experience when I was fixing my bosses esl-63’s where the voice sounded like it was behind the speaker.
The reason is at least in part because they both radiate a simple segment of a sphere, from the same location in time and space, over a wide bandwidth. When that happens, what arrives at the right and left ear from one speaker is “as if” it was from the same source. One can only localize the depth location of a speaker IF there is enough difference in the speakers radiation between what arrives at your right and left ears. If what arrives is similar enough, then we hear the clues in the recording more strongly.
In large scale, the lack of an interference pattern is VERY audible, particularly as the wind blows, hardly anything happens to the sound and the frequency response shape is essentially identical everywhere (is different everywhere with an interference pattern).
I have noticed too that small full range drivers on a large flat baffle produce a similar effect, over much of their range, they radiate as a simple spherical segment and not an interference pattern.
It is interesting that Barleywaters arrangement being so small it would also radiate as simple spherical pattern as well up to tweeter directivity.
While it is known speakers sound better when you deal with edge diffraction and such, I am not sure people realize that is also one way to produce an interference pattern, re-radiation where there is an abrupt acoustic impedance change. Horn are the ultimate extension of proving a controlled boundary out to a large dimension where a wave can freely propagate at the same angle (as defined by Don Keeles pattern loss thumb rule).
I believe what is boils down to at least spatially speaking, is that with only one speaker on, the more similar the signal reaching both ears, the less information is carried about the speaker as the source in physical space. When the speaker radiates little about itself, then it’s harder to hear distance to the speaker, the more strong the mono phantom image is / the less able you hear a right and left sources and when the source identity drops away, then you have more information arriving intact from the recording..
If you have headphones, here is video a customer of ours took of two pairs of speakers he owns, the first part shows the source identity thing.
Danley Sound Labs SM80 vs. d&b Q7 - YouTube
They have a very similar frequency response (one is eq to match the other ) so the spectral content / balance is very similar. When he switches back and forth, you can hear a pretty clear difference in how it sounds but it isn’t anything to do with frequency response but how the two sources radiate in time and space. What you hear has to do with the stereo image and the source identity although I am not sure I could explain it if you haven’t heard it.
Best,
Tom Danley
This is an interesting hypothesis but shouldn't frequency dependent interaural intensity differences caused by lobing provide less coherent localization cues, therefore making it harder to localize the speaker?
Hi Pano, markus
Hey that’s right you have heard them at the AES thing, I keep thinking no one here has heard them.
It’s too bad they had the seating like that with an isle in the sweet spot for the image in the recordings at the end. Glad I don’t have to deal with acoustics in exotic spaces after the architect has decided what it has to look like, sheesh a nightmare that guy from Turkey had to deal with.
How have you been, did you get much of “that weather” that was blowing around that edge of the country?
Hi Markus
What I described is what it looks like to me right now.
My desire to make sense of this began, like the idea for the Unity and Synergy horns, with something Don Davis at Synaudcon said in the late 80’s.
My world was below 100Hz then and after making a small servomotor driven subwoofer called a Contra bass, Don asked to play with one (weird driver being driven by a rotary motor and all).
At that time he was also playing with what they called “in the ear” recordings. With that, tiny microphones connected with real small Teflon tubes were used. The tube was carefully snaked into your ear so that what the mic heard was the pressure next to your ear drum.
Testing with volunteers shows that the outside of your ear had a large variety of effects on the sound getting into your ear and many depended on incoming angle and height etc, ALL direction of source things. Your head causes some effects because it is an obstruction but the shape of your ear is a large factor too.
Anyway, Don liked the subwoofer and said having low frequencies made more difference than he had guessed and invited me to hear the recordings with the Contrabass at his house. You sat in a chair (only room for one) and the speakers were on the floor facing up from each side.
The recording he played was at the Indy 500 time trials, the volunteer was walking around the pits and in the not too far distance was the occasional car zooming by. It made the hair on the back of my neck stand up, it gave me goose bumps, I have never heard anything that realistic and unfortunately impractical.
I hate to think it has taken this long to put together an explanation but it essentially has to be that every thing we measure when we look at the inner ear pressure response, things that make the ear look like it’s broken if it were a speaker, are the things we have learned over a lifetime that tell us where the sound is coming from. When a sound goes from low to high in position, we don’t hear the series of notches and artifacts moving around, we hear the source moving around.
We don’t hear that our ears response is nothing like flat, similarly, it is the only thing we know. The in the ear recording captured all of those aberrations and presented them to the listener from an angle from which we have few learned clues or acuity (down).
So, imagine two situations, one where with your eyes closed you can easily estimate within a few feet or less, how far away the speaker is. The other a situation where you can hear the height and direction but more strongly sounds far away or close, depending on the recording. With little source identity, a right and left speaker producing the same signal, one gets a strong mono phantom and very little sense of a right and left source.
All things like response and distortion being equal, the ability to localize the loudspeakers position in depth essentially has to be from differences between what is reaching the two ears in one case but not the other because that’s how we hear how far away or where a source of sound is..
Or at least that is more or less what I think as of right now haha.
Best,
Tom Danley
Hey that’s right you have heard them at the AES thing, I keep thinking no one here has heard them.
It’s too bad they had the seating like that with an isle in the sweet spot for the image in the recordings at the end. Glad I don’t have to deal with acoustics in exotic spaces after the architect has decided what it has to look like, sheesh a nightmare that guy from Turkey had to deal with.
How have you been, did you get much of “that weather” that was blowing around that edge of the country?
Hi Markus
What I described is what it looks like to me right now.
My desire to make sense of this began, like the idea for the Unity and Synergy horns, with something Don Davis at Synaudcon said in the late 80’s.
My world was below 100Hz then and after making a small servomotor driven subwoofer called a Contra bass, Don asked to play with one (weird driver being driven by a rotary motor and all).
At that time he was also playing with what they called “in the ear” recordings. With that, tiny microphones connected with real small Teflon tubes were used. The tube was carefully snaked into your ear so that what the mic heard was the pressure next to your ear drum.
Testing with volunteers shows that the outside of your ear had a large variety of effects on the sound getting into your ear and many depended on incoming angle and height etc, ALL direction of source things. Your head causes some effects because it is an obstruction but the shape of your ear is a large factor too.
Anyway, Don liked the subwoofer and said having low frequencies made more difference than he had guessed and invited me to hear the recordings with the Contrabass at his house. You sat in a chair (only room for one) and the speakers were on the floor facing up from each side.
The recording he played was at the Indy 500 time trials, the volunteer was walking around the pits and in the not too far distance was the occasional car zooming by. It made the hair on the back of my neck stand up, it gave me goose bumps, I have never heard anything that realistic and unfortunately impractical.
I hate to think it has taken this long to put together an explanation but it essentially has to be that every thing we measure when we look at the inner ear pressure response, things that make the ear look like it’s broken if it were a speaker, are the things we have learned over a lifetime that tell us where the sound is coming from. When a sound goes from low to high in position, we don’t hear the series of notches and artifacts moving around, we hear the source moving around.
We don’t hear that our ears response is nothing like flat, similarly, it is the only thing we know. The in the ear recording captured all of those aberrations and presented them to the listener from an angle from which we have few learned clues or acuity (down).
So, imagine two situations, one where with your eyes closed you can easily estimate within a few feet or less, how far away the speaker is. The other a situation where you can hear the height and direction but more strongly sounds far away or close, depending on the recording. With little source identity, a right and left speaker producing the same signal, one gets a strong mono phantom and very little sense of a right and left source.
All things like response and distortion being equal, the ability to localize the loudspeakers position in depth essentially has to be from differences between what is reaching the two ears in one case but not the other because that’s how we hear how far away or where a source of sound is..
Or at least that is more or less what I think as of right now haha.
Best,
Tom Danley
I've sort of heard them (just my own diy builds based on stuff gleaned from patents and forum posts, still yet to hear the real thing).
I think two other things that limits the ability to tell "how far away" Synergy horns are would be the well-controlled directivity and point-source behavior causing a simpler effect. Take away the directivity, and the mix of direct vs reflected sound becomes dependent on distance from the speaker -- you get a higher fraction of reflected and late reflected the further you are. But with controlled narrower directivity the change with distance is quite a bit less, more of a headphone effect. Similarly, the effect of lobeing with non-point source speakers cause a difference in the reflected sound response shapes experienced depending on where you are and how the lobes are reflecting around the room; make it point-source and it's pretty much the same character everywhere whether the speaker is close or far.
How that works into getting the rather spooky phantom center these things present isn't explained by those simpler effects, though. But probably the horns causing you to be less aware of just where the speakers are in any dimension including distance (and not having your brain fighting so hard to ignore knowing where they are!) helps in letting the sound seem as if it is from other places - including the center. I similarly get outside the speaker sounds at times (wondering whether the source was surround in some cases), helped, I think, by the same lack of localization of the speakers.
How so? I'd expect such a quasi omnipolar design to create lots of reflections. Did you listen to them outside?
Sure but why would a speaker with extreme lobing provide more cues revealing its position than a speaker with a smooth polar pattern? It should be the other way around.
Just listened to the video over headphones. The sound stage of the other speaker is different from the SM80. Voices are not centered but shifted to the right.
The different tonality is probably caused by differences in power response which can be heard in the 90 degrees off-axis example.
I'd like to hear one of your speakers in my room. Is there someone in Zurich, Switzerland who has them?
What is easier to localize, a source with widely varying radiation pattern or a source with an omnidirectional radiation pattern?
An interesting hypothesis is that stereo works best when speakers are easier to localize, so the brain gets more time to process other cues.
Last edited:
You may ask for too much, Danley is not that popular in EU yet. I know someone in Budapest who's got the whole line, and someone else in Graz selling his set of 3 SH-50. Both are too far for me..
Hi Markus, Bill W (I have not forgotten)
Understand, this is what it looks like to me, not a widely accepted explanation.
“What it sounds like to us” is a presentation our brain and hearing system compose for us. We have spent all of our lives exposed to those inputs, learning what they mean and over time and to varying degrees, we learn to be able to “hear” a great deal about the sound direction, height and so on. All of those abilities are a result of the head as an obstruction and the shape of your outer ear and inner passages and the changes they cause to sound as a function of incoming angle and height etc.
In physics, it normally requires one more frame of reference than the dimension you with to describe.
By that I mean, if you have two frames of reference, you have the ability to describe one dimension and so theoretically, our two ears could only resolve one place, left to right. We get around that limitation and can hear height, because of the changes your outer ear causes as a function of height.
Again, we can’t hear those changes like they appear in a measurement, we have learned they correspond to height instead.
If curious about how strong the learned part of hearing is, Google “the mcgurke effect” from a cool BBC documentary on the senses.
If one is outside, for one person, then the loudspeakers directivity doesn’t matter for this, all things equal, you could not hear the difference between an omni and a speaker with a Q of 50. What matters is the difference in the sound reaching the right and left ear from one speaker compared to that difference that would exist IF it had all come from one point in space..
Playing a soft voice , If it came from one point in space (not like a live voice ), it would be a simple smooth expanding spherical segment, like a bubble and if we face the source level, it sounds like it is in front of you, but you can’t tell how far away it is because what reaches both ears is essentially identical.
If you faced a live person talking, there are significant differences between the right and left ear that let you guess their distance or “hear” where their mouth is.
In several other threads I have suggested an experiment where the curious can hear this kind of effect.
Get a pair of the small fostex full range drivers and mount them on some large flat baffles, say 2-3 feet square minimum. With that, the driver will radiate most of a hemisphere up to where the cone has directivity (the hf is all breakup mode). Arrange these so that you are sitting closer than normal, say half the distance and preferably towards the center of the room to minimize close reflections.
You can eq them some if you want and sitting closer / being towards the room center, lets you be in the near field. The big flat baffle, lets the sound to expand without perturbation with the “bubble” pressure gradient perpendicular to the horn wall, like in a conical horn.
These are not perfect but I have heard these several times on large baffles and have been very impressed by the imaging / lack of source identity each time.
A Manger and ESL-63 are also speakers where I have heard the apparent source of sound in physical depth not be strongly associated with the loudspeaker. Both of these also radiate a segment of a sphere over a wide bandwidth.
There are of course many things which effect how a speaker sounds but how the speaker radiates IS one that is i think very under appreciated and profoundly connected to stereo imaging. All it requires to hear this effect is a source that is acoustically small so it has no real directivity and no diffraction how it’s mounted or like with the manger, the esl-63 or Synergy horn, to radiate as if it had come from a single tiny source.
I suppose if one tried this on a desktop scale, one could eq a mid dome and extend the operation higher and lower than normal. Remember we are talking about hearing the radiation, not that these are ideal solutions. I believe this aspect is the root of why some people swear by crossover-less full range drivers.
Best
Tom
Understand, this is what it looks like to me, not a widely accepted explanation.
“What it sounds like to us” is a presentation our brain and hearing system compose for us. We have spent all of our lives exposed to those inputs, learning what they mean and over time and to varying degrees, we learn to be able to “hear” a great deal about the sound direction, height and so on. All of those abilities are a result of the head as an obstruction and the shape of your outer ear and inner passages and the changes they cause to sound as a function of incoming angle and height etc.
In physics, it normally requires one more frame of reference than the dimension you with to describe.
By that I mean, if you have two frames of reference, you have the ability to describe one dimension and so theoretically, our two ears could only resolve one place, left to right. We get around that limitation and can hear height, because of the changes your outer ear causes as a function of height.
Again, we can’t hear those changes like they appear in a measurement, we have learned they correspond to height instead.
If curious about how strong the learned part of hearing is, Google “the mcgurke effect” from a cool BBC documentary on the senses.
If one is outside, for one person, then the loudspeakers directivity doesn’t matter for this, all things equal, you could not hear the difference between an omni and a speaker with a Q of 50. What matters is the difference in the sound reaching the right and left ear from one speaker compared to that difference that would exist IF it had all come from one point in space..
Playing a soft voice , If it came from one point in space (not like a live voice ), it would be a simple smooth expanding spherical segment, like a bubble and if we face the source level, it sounds like it is in front of you, but you can’t tell how far away it is because what reaches both ears is essentially identical.
If you faced a live person talking, there are significant differences between the right and left ear that let you guess their distance or “hear” where their mouth is.
In several other threads I have suggested an experiment where the curious can hear this kind of effect.
Get a pair of the small fostex full range drivers and mount them on some large flat baffles, say 2-3 feet square minimum. With that, the driver will radiate most of a hemisphere up to where the cone has directivity (the hf is all breakup mode). Arrange these so that you are sitting closer than normal, say half the distance and preferably towards the center of the room to minimize close reflections.
You can eq them some if you want and sitting closer / being towards the room center, lets you be in the near field. The big flat baffle, lets the sound to expand without perturbation with the “bubble” pressure gradient perpendicular to the horn wall, like in a conical horn.
These are not perfect but I have heard these several times on large baffles and have been very impressed by the imaging / lack of source identity each time.
A Manger and ESL-63 are also speakers where I have heard the apparent source of sound in physical depth not be strongly associated with the loudspeaker. Both of these also radiate a segment of a sphere over a wide bandwidth.
There are of course many things which effect how a speaker sounds but how the speaker radiates IS one that is i think very under appreciated and profoundly connected to stereo imaging. All it requires to hear this effect is a source that is acoustically small so it has no real directivity and no diffraction how it’s mounted or like with the manger, the esl-63 or Synergy horn, to radiate as if it had come from a single tiny source.
I suppose if one tried this on a desktop scale, one could eq a mid dome and extend the operation higher and lower than normal. Remember we are talking about hearing the radiation, not that these are ideal solutions. I believe this aspect is the root of why some people swear by crossover-less full range drivers.
Best
Tom
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.