The perfect speaker is room dependent - wide vs. narrow directivity and more

[dasdoing]

This can be accomplished by using sufficently directional speakers and toeing them in aggressively, such that the first strong lateral reflection for the left-hand speaker is the long across-the-room bounce off the right-hand wall, and vice versa.

that's the worst reflection.
we also need at least 50ms of interval for it to be percieved as a second source. that would be an extra travel distance of 17m

 

[dasdoing]

I've found out perception of stereo image shifts considerably at some particular listening distance. Perception seems to shift from 2D and slightly far and hazy sound to more enveloping and clear 3D kind of presentation when listening at close enough distance, with small enough listening triangle.

I percieve it like this, and I have seen many sound engeniers describe it like this over the years. you create an equilateral triangle and the ambience isn't really there yet. but as soon as you move inside, doesn't need to be much, the ilusion is there. So many, and I, will setup the system so that the triangle closes just behind our head

 

[tmuikku]

I bet you can shrink the base width and adjust toe-in if you wish, these are important for the stereo image size and how the stereo image feels. But the effect, the transition between 2D/3D, seems to be mostly due to distance from ear to speaker, D/R ratio. Going inside equilateral listening triangle reduces this distance as well. But, this is just my experience and reasoned based on that, your explanation might be more correct. Its the same thing either way, same perceived phenomenon no matter the explanation

I have my setup now roughly 45 deg angle, basically what you describe, listening spot is inside equilateral listening triangle. Sometimes this is too wide feel especially with hard panned instruments, like drum toms panned to different sides, just weird feel I have adjusted the setup so that on the listening spot I can lean back to get sound seem further away and bit narrower, easier with the fancy panned stuff. Or lean forward to zoom inside the sound so to speak. This seems to work nicely, comfortable and interactive.

I can get the 3D feel with equilateral listening triangle as well, but sometimes this seems too narrow, as there is no aid from room to widen the stereo image when listening at the close proximity.

 

[dasdoing]

I bet you can shrink the base width and adjust toe-in if you wish, these are important for the stereo image size and how the stereo image feels. But the effect, the transition between 2D/3D, seems to be mostly due to distance from ear to speaker, D/R ratio. Going inside equilateral listening triangle reduces this distance as well. But, this is just my experience and reasoned based on that, your explanation might be more correct. Its the same thing either way, same perceived phenomenon no matter the explanation

I have my setup now roughly 45 deg angle, basically what you describe, listening spot is inside equilateral listening triangle. Sometimes this is too wide feel especially with hard panned instruments, like drum toms panned to different sides, just weird feel I have adjusted the setup so that on the listening spot I can lean back to get sound seem further away and bit narrower, easier with the fancy panned stuff. Or lean forward to zoom inside the sound so to speak. This seems to work nicely, comfortable and interactive.

I can get the 3D feel with equilateral listening triangle as well, but sometimes this seems too narrow, as there is no aid from room to widen the stereo image when listening at the close proximity.

I can't share your experience. I have used various sizes of triangles over the years, and the distance doesn't seam to be a factor. what will happen in an untreated room is that the bigger the distance to more the ilusion is disturbed by reflections

 

[tmuikku]

Yep, perhaps there is misunderstanding as that is the same thing I'm trying to say? the bigger the triangle, the further away you are from the speakers and the more direct sound attenuates while room sound stays pretty much at same level regardless of where speakers and you are in the room. This means the further one is from speakers D/R ratio goes down, and as you say the louder the reflections are and disturb the illusion.

The shift in perceived stereo impression seems to happen at certain distance (and toe-in), when close enough to speakers so when D/R is high enough on some critical bandwidth. Both speaker directivity and room acoustic properties are frequency dependent: bass is always only room sound, while highs are always about direct sound with room sound coming in much later, but some bandwidth between seems to be important and responsible for shift in stereo illusion. Griesinger mentions 700-4kHz bandwidth to contain important harmonics, at least for speech.

According to Griesinger hearing system is able to pick out important sounds from all the sounds in surroundings by detecting amplitude peaks caused by harmonics. Basically, when phases of the harmonics are intact they make huge amplitude peaks every cycle of fundamental, which provides high enough signal-to-noise ratio for that particular sound and brain is able to pick it out from the noise. When it does so, the important sound gets its own neural stream while the background (room sounds) get another, clarity and envelopment. If listening too far, the noise is too great / harmonics scrambled and brain is not able to pick out the direct sound but just hears one sound, direct sound mixed in with all the other sounds in the room, there is no clarity nor envelopment

 

[dasdoing]

Yep, perhaps there is misunderstanding as that is the same thing I'm trying to say? the bigger the triangle, the further away you are from the speakers and the more direct sound attenuates while room sound stays pretty much at same level regardless of where speakers and you are in the room. This means the further one is from speakers D/R ratio goes down, and as you say the louder the reflections are and disturb the illusion.

The shift in impression seems to happen at certain distance (and toe-in), when close enough speakers, when D/R is high enough on some critical bandwidth. Both speaker directivity and room acoustic properties are frequency dependent: bass is always only room sound, while highs are always about direct sound with room sound coming in much later, but some bandwidth between seems to be important and responsible for shift in stereo illusion.

Griesinger writes its roughly around 1khz, perhaps 700-4kHz that is responsible for it at least with speech. Brain locks in to the harmonics of sounds, and the harmonics are are mostly on this bandwidth. The lock in happens when phases of the harmonics are intact which makes huge amplitude peaks every cycle of fundamental, which provides high enough signal-to-noise ratio for that particular sound and brain is able to pick it out as important close sound from all the surrounding sounds, which is basically just noise. When it does so, there is clarity and envelopment, the close sound.

got you.
yea, having a strong direct sound is crucial for the imagaing in general.

 

[Duke]

that [the first contralateral reflection]'s the worst reflection.

That has not been my experience; quite the contrary, but then I'm using speakers specifically designed for this kind of set-up which may make a difference.

we also need at least 50ms of interval for it to be percieved as a second source. that would be an extra travel distance of 17m

The idea is NOT for the first contralateral reflection to be perceived as a secondary sound source; the idea is to delay its arrival and de-correlate it relative to the first-arrival sound, which will have arrived at the other ear. This way it contributes to spaciousness without the arguably negative side-effects of an early-arriving ipsilateral reflection.

Earl Geddes on the subject:

"The earlier and the greater in level the first room reflections are, the worse they are. This aspect of sound perception is controversial. Some believe that all reflections are good because they increase the listeners' feeling of space – they increase the spaciousness of the sound. While it is certainly true that all reflections add to spaciousness, the very early ones (< 10 ms.) do so at the sake of imaging and coloration. There is no contention that reflections > 20 ms are positive and perceived as early reverberation and acoustic spaciousness within the space. In small rooms, the first reflections from an arbitrary source, mainly omnidirectional, will never occur later than 10-20 ms (basically this is the definition of a small room), hence the first reflections in small rooms must be thought of as a serious problem that causes coloration and image blurring. These reflections must be considered in the [loudspeaker] design and should be also be considered in the room as well.

"Reflections become less of a problem as coloration and image shift at lower frequencies. Below about 500 Hz. early reflections are not as much of an issue. The ear has a longer integration time at lower frequencies and it has a poorer ability to localize resulting in a lower sensitivity to early reflections. Image localization is strongly weighted towards the higher frequencies.

"A reflected signal that arrives at the opposite ear from the direct sound is less perceptible as coloration and image shift than if both signals arrive at the same ear. This is because of head shadowing above about 500 Hz and the fact that our ears can process signals between them. When the two signals arrive at the same ear, the signals are physically merged in space even before they enter the ear and no amount of auditory processing can separate them. When these signals arrive at different ears, the auditory processing system can diminish the adverse effects of these early reflections through cognitive processing between the ears."

And a short video clip:

 

[digitalfrost]

So many, and I, will setup the system so that the triangle closes just behind our head

That's also how I have it.

 

[dasdoing]

That has not been my experience; quite the contrary, but then I'm using speakers specifically designed for this kind of set-up which may make a difference.



The idea is NOT for the first contralateral reflection to be perceived as a secondary sound source; the idea is to delay its arrival and de-correlate it relative to the first-arrival sound, which will have arrived at the other ear. This way it contributes to spaciousness without the arguably negative side-effects of an early-arriving ipsilateral reflection.

Earl Geddes on the subject:

"The earlier and the greater in level the first room reflections are, the worse they are. This aspect of sound perception is controversial. Some believe that all reflections are good because they increase the listeners' feeling of space – they increase the spaciousness of the sound. While it is certainly true that all reflections add to spaciousness, the very early ones (< 10 ms.) do so at the sake of imaging and coloration. There is no contention that reflections > 20 ms are positive and perceived as early reverberation and acoustic spaciousness within the space. In small rooms, the first reflections from an arbitrary source, mainly omnidirectional, will never occur later than 10-20 ms (basically this is the definition of a small room), hence the first reflections in small rooms must be thought of as a serious problem that causes coloration and image blurring. These reflections must be considered in the [loudspeaker] design and should be also be considered in the room as well.

"Reflections become less of a problem as coloration and image shift at lower frequencies. Below about 500 Hz. early reflections are not as much of an issue. The ear has a longer integration time at lower frequencies and it has a poorer ability to localize resulting in a lower sensitivity to early reflections. Image localization is strongly weighted towards the higher frequencies.

"A reflected signal that arrives at the opposite ear from the direct sound is less perceptible as coloration and image shift than if both signals arrive at the same ear. This is because of head shadowing above about 500 Hz and the fact that our ears can process signals between them. When the two signals arrive at the same ear, the signals are physically merged in space even before they enter the ear and no amount of auditory processing can separate them. When these signals arrive at different ears, the auditory processing system can diminish the adverse effects of these early reflections through cognitive processing between the ears."

And a short video clip:

the 20ms gap is mentioned very often, and I can totally believe in it, but it's still an extra flytime of almost 7 meters

 

[Duke]

the 20ms gap is mentioned very often, and I can totally believe in it, but it's still an extra flytime of almost 7 meters

A full 20 milliseconds free of reflections in the hozontal plane is not practical in most domestic rooms. I recall reading a paper (probably an AES paper) where a 15 milliseconds reflection-free interval was proposed as a standard, or maybe it already was the standard, but I don't remember anything else about the paper so I don't think I could find it again.

Earl Geddes shoots for 10 milliseconds and so do I, with Earl basing that figure on Griesinger's work and on gammatone filters. Initially I based my 10 milliseconds target on sighted listening tests varying the distance of dipole speakers from the "front" wall so that I could make recommendations to customers, but now I cite Earl Geddes and David Griesinger because they actually have credibility. Who's going to believe a dealer saying "this is what I heard", right??

In this context my recollection is that Siegfried Linkwitz indicated 6 milliseconds was about the minimum recommended reflection-free interval, which also corresponded with my sighted listening tests and found its way into the recommendations I make to my customers.

Having compared optimizing for reflection arrival direction (60 degrees to the left and right of center) versus optimizing for reflection arrival time (with 10 milliseconds being the target), ime arrival time is what matters the most. I say this within the context of a two-channel stereo system.

 

[Descartes]

Did you test it in a real world set up? Envelopment is hard to create even with that many speakers. The recording technique is also essential. Which 3D audio track captured envelopment well for such a set up?

Yes and yes I heard two systems one was Steinway Lygdorf the other a JBL Synthesis system both were very impressive just extremely expensive that was quite amazing!

Try some of 2L recording on BluRay Audio

 

[dasdoing]

A full 20 milliseconds free of reflections in the hozontal plane is not practical in most domestic rooms. I recall reading a paper (probably an AES paper) where a 15 milliseconds reflection-free interval was proposed as a standard, or maybe it already was the standard, but I don't remember anything else about the paper so I don't think I could find it again.

Earl Geddes shoots for 10 milliseconds and so do I, with Earl basing that figure on Griesinger's work and on gammatone filters. Initially I based my 10 milliseconds target on sighted listening tests varying the distance of dipole speakers from the "front" wall so that I could make recommendations to customers, but now I cite Earl Geddes and David Griesinger because they actually have credibility. Who's going to believe a dealer saying "this is what I heard", right??

In this context my recollection is that Siegfried Linkwitz indicated 6 milliseconds was about the minimum recommended reflection-free interval, which also corresponded with my sighted listening tests and found its way into the recommendations I make to my customers.

Having compared optimizing for reflection arrival direction (60 degrees to the left and right of center) versus optimizing for reflection arrival time (with 10 milliseconds being the target), ime arrival time is what matters the most. I say this within the context of a two-channel stereo system.

in studios a lot of experimentation has been done with controled reflections. google "haas-kicker" for example. these concepts were all abandoned because of all the extra comb filtering it causes.
the better solution for all this is to use diffusion. you get a nice delayed tail, instead of a "hard reflection"