Bass direction is audible

[Flaesh]

I would love to hear ASR's opinion about this.

Visualizing How Different Loudspeaker LF Directivity Patterns Couple to Room Modes

This opening post is to demonstrate how different loudspeaker low frequency directvity patterns couple to our room modes. If you are unfamiliar with what room modes are, please refer to @DonH56's excellent thread for an in-depth introduction...

www.audiosciencereview.com

I don't know why NTK didn't provide this link))

 

[Flaesh]

*Bass (D or S) array radiates in one direction. Plain travelling wave. This is not a true monopole. But if desired , Di can be considered infinite.

 

[gnarly]

I had a debate with a friend about whether cardioid subs make any sense in living rooms. I know that they make sense outdoors or in large concert halls. But I am sceptical if they make any difference in a room where the longest dimension is shorter than a bass wavelength. Am I wrong? I would love to hear ASR's opinion about this.

My take is that cardioid sub technology as used by large scale live sound, is pretty much inapplicable for home use. The cardioid sub arrays are almost invariably active, big, and need some space around them to breathe correctly. They also tend to focus their attenuation to a rather specific polar direction, and over a rather narrow freq range.

The passive cardioid route however, may make good sense for home.

Here's one prosound manufacturer who's touting it for installs.

(Fulcrum Acoustic has a very high reputation within the prosound industry, both for product and technological innovations.)
Appears they are getting about -10dB attenuation to the rear over a fairly wide freq range, that also extends somewhat to the sides.

 

[audiofooled]

Visualizing How Different Loudspeaker LF Directivity Patterns Couple to Room Modes

This opening post is to demonstrate how different loudspeaker low frequency directvity patterns couple to our room modes. If you are unfamiliar with what room modes are, please refer to @DonH56's excellent thread for an in-depth introduction...

www.audiosciencereview.com

I don't know why NTK didn't provide this link))

I think It doesn't hurt visualizing phase, also time delays over frequency:

Part 2:

And yet another link to David Griesinger's paper on Loudspeaker and listener positions for optimal low-frequency spatial reproduction in listening rooms:

http://www.davidgriesinger.com/asa05.pdf

 

[audiofooled]

As far as localization goes, auditory mechanism is sensitive to signals with sharp envelope onset. Bass transients are, more often than not, containing wide range of frequencies starting in several kHz region, which is quick, sweeping down to the low frequencies which are slower than slow. To illustrate, in attachment you may find a video of gravity vectors I previously posted, with a caveat of adding a super slow motion at the beginning of a transient.
You may get the picture of orders of magnitude quicker solid floor bounce at the start of the transient (only a couple of single frames up and down) in comparison to a very, very slow displacement from left to right and back to the state of equilibrium, which would take minutes I suppose and be very boring.

Now, floor is a very large and solid plane and if not taken care of, it can be a source of distortion. Just my 2c, for people who have subs on the floor. Plenty of chance to localize the sub as the source, and not the source material cues.

 

[Chromatischism]

Now, floor is a very large and solid plane and if not taken care of, it can be a source of distortion. Just my 2c, for people who have subs on the floor.

What do you mean by that?

 

[audiofooled]

What do you mean by that?

As I said earlier, having system that is capable of not compressing the signals with sharp onset, which have higher crest factor than sinewaves, floor bounce is faster than sound. This plane wave crest energy is upwards, and by the time the low frequencies are starting to get reproduced, it had already induced a feedback and resonance into the loudspeaker cabinet (if it's coupled hard to the floor). If you have a system that is capable of such dynamics, it's very, very tangible. Propagating through the floor, It's lifting the entire room, furniture and small objects along with it. Having a tea spoon in an empty cup put in the kitchen sink is easily audible and your wife may hear it and see it vibrate.

Although the frequencies are higher and not a job of a subwoofer to reproduce, it can vibrate the enclosure all the same. Funny thing is that it's usually not audible until you get rid of it and realize that it was.

Dealing with this is not at all simple (decoupling). There are commercial products (pucks of dampening materials-isolation feet) which may, or may not work, depending on your situation.

I've posted this a while ago, but here it is (what happens on the ground) on a system that is so powerful and big that it has directivity in low frequencies. Slow it down to 0.25 and you may see that during transients ground moves and waves through it propagate faster than the sound reaches the camera. Being an array of subs, it's impulse response is stretched in time, creating waves on the ground that are visible when lifting the dust particles:

 

[Chromatischism]

As I said earlier, having system that is capable of not compressing the signals with sharp onset, which have higher crest factor than sinewaves, floor bounce is faster than sound. This plane wave crest energy is upwards, and by the time the low frequencies are starting to get reproduced, it had already induced a feedback and resonance into the loudspeaker cabinet (if it's coupled hard to the floor). If you have a system that is capable of such dynamics, it's very, very tangible. Propagating through the floor, It's lifting the entire room, furniture and small objects along with it. Having a tea spoon in an empty cup put in the kitchen sink is easily audible and your wife may hear it and see it vibrate.

Although the frequencies are higher and not a job of a subwoofer to reproduce, it can vibrate the enclosure all the same. Funny thing is that it's usually not audible until you get rid of it and realize that it was.

Dealing with this is not at all simple (decoupling). There are commercial products (pucks of dampening materials-isolation feet) which may, or may not work, depending on your situation.

Physical resonances transferred from transducers through solid objects are not the same as the sound energy coming from the subwoofer. Decoupling the subwoofer, as you say, will prove that.

I've posted this a while ago, but here it is (what happens on the ground) on a system that is so powerful and big that it has directivity in low frequencies. Slow it down to 0.25 and you may see that during transients ground moves and waves through it propagate faster than the sound reaches the camera. Being an array of subs, it's impulse response is stretched in time, creating waves on the ground that are visible when lifting the dust particles:

I think different concepts are being mixed up here. You can create directivity with subwoofer arrays for outdoor venues by their physical alignment and their time alignment. That is how they achieve some semblance of sound containment. This is not really applicable at home.

But you still haven't answered the question. What do you mean by "taking care of the floor"?

 

[audiofooled]

Physical resonances transferred from transducers through solid objects are not the same as the sound energy coming from the subwoofer. Decoupling the subwoofer, as you say, will prove that.

I have no problems with resonances from transducers or the boxes, not from the mains, not from the sub. It was the floor that was creating resonances and decoupling isn't as easy as it seems. It will depend on Q of resonances and it can be such that the decoupling products are not one size fits all. Even though they are usually advertised as a guaranteed improvement (as all other audiophile tweaks), they may partially work, or not.

I think different concepts are being mixed up here. You can create directivity with subwoofer arrays for outdoor venues by their physical alignment and their time alignment. That is how they achieve some semblance of sound containment. This is not really applicable at home.

People create sub arrays at home. It works. It's just not always practical.

But you still haven't answered the question. What do you mean by "taking care of the floor"?

Decoupling in itself is not a complete solution, even if done right. Room is a part of the system and floor is a part of the room. Carpets won't help. Read Griesinger's paper.

 

[audiofooled]

I feel like sharing some more of my insights about this topic, with regards to waveforms you may find in music. There are various saturation tools for adding distortion and manipulating it, making complex waveforms and balancing them so that they don't sound bad even when summed to mono. For this there are tools for measuring correlation in between channels as to avoid phasing issues during summation. Finally, tools that make transient attacks stand out by listening to frequencies that are sharing the same bandwidth and turn down sustained ones during transients.

Here's an interesting video about ways to use saturation products in the mix:

At the end of the video there's a link to another one titled "You don't understand saturation", which I highly recommend as well.

So, in a nutshell, music is rarely pure tones, because people like distortion.

I would like to share an example of a track where you may observe the utilization of such tools to get wide, stereo and powerful bass. It is titled quite conveniently "Realm Shift" (I suppose @gnarly could shift the entire neighborhood with his setup) :

This is some serious bass. I would like to focus on a 10 seconds part of the track starting at about 1:07. If we look into it, you may see that this is anything but pure sinewaves, also stereo bass:

It is also pulsating within some low frequency sidebands, so I suspect complex phase relations as well. For this part of the track, bass does not go lower than 40Hz.

Reproduced on my system, this gives complex gradient experienced at my MLP. In trying to remotely describe this, I've set the device next to me, so it is slanted and positioned this way:

It is placed horizontally, with the upper edge of the screen (straight up arrow) pointed towards the system. It is also placed on my left, so the right arrow is pointing at me. This is 2D if you look at it, but in fact it is 3D as the upward movement is displayed as a complex product of the both Y(red) and X(blue) axis, although in reality it is not displaced from left to right. Screenshot actually shows neutral position.

In attachment you may find another real time recording, but this time I did my best not to drop any frames if possible (it's about 57fps), and I also mixed in some of my room sound together with the original track, for a better audio-visual clue and higher sound quality. The original track is dry, so I thought a bit of reverberation field from my room helps in sharing my experience. And vice versa, my in room recording wouldn't contain low frequencies so I had to mix them in from the original track. I did my best to achieve at least some balance.

I also made it in .mov format, so if you are using Quicktime player, you may drag and observe individual frames, also manipulate speed.

What you may observe is that frequencies are having a life of their own with regards to positions along the axis, also that they are rather abruptly stopped and repositioned during transients, no matter the amplitude they were in. My setup is such that this gives me cues such as that anything that is transient in nature, it is perceived in front, close to the system. But gradient is such that it well correlates with what you see on the video, so the lower the sustained frequency it is localized behind me, with higher frequencies localized at my position or in front. This track is stereo bass, so there's also perceived width and pulses of more or less width, which sadly I cannot explain. Far too complex and beyond my paygrade.

I hope you people find this interesting. Or perhaps not. Comments are welcome.

 

[TheZebraKilledDarwin]

@audiofooled
I love to read sermons from laymen, how they would do brain surgery...
You are talking about a topic, you lack the mathematical basics to understand signals and waves and therefore acoustics. Please stop!

 

[audiofooled]

@audiofooled
I love to read sermons from laymen, how they would do brain surgery...
You are talking about a topic, you lack the mathematical basics to understand signals and waves and therefore acoustics. Please stop!

You're right. Mathematically proving that sugar tastes sweet would be an arduous task. Very well then. I wish you the best of luck in your future endeavors.

 

[Emlin]

You're right. Mathematically proving that sugar tastes sweet would be an arduous task. Very well then. I wish you the best of luck in your future endeavors.

Why so bitter?

 

[TheZebraKilledDarwin]

You're right. Mathematically proving that sugar tastes sweet would be an arduous task. Very well then. I wish you the best of luck in your future endeavors.

But you are not saying that sugar tastes sweet and good for YOU, you are trying to make conclusions about the chemistry and physics of sugar without having the formal education to understand chemical formulas.
Talking about distortion of signals without understanding the Fourier-theorem but making conclusions about the technicals of distorted signals, is like someone not knowing what makes the motor of a car move, but because he likes the car he wants to play mechanic and tries to explain how to repair a car without learning the stuff necessary to understand it.
Common sense should tell you, that it will not work. First you need to learn and understand the basics, then you can talk.
That's one of the wise reasons, why children had been thought to shut up, when adults were talking.

 

[audiofooled]

But you are not saying that sugar tastes sweet and good for YOU, you are trying to make conclusions about the chemistry and physics of sugar without having the formal education to understand chemical formulas.
Talking about distortion of signals without understanding the Fourier-theorem but making conclusions about the technicals of distorted signals, is like someone not knowing what makes the motor of a car move, but because he likes the car he wants to play mechanic and tries to explain how to repair a car without learning the stuff necessary to understand it.
Common sense should tell you, that it will not work. First you need to learn and understand the basics, then you can talk.
That's one of the wise reasons, why children had been thought to shut up, when adults were talking.

Oh please, I do not have time for this. But I guess I'm guilty for trying to keep things simple.

 

[EJ3]

I don't think the anechoic perceptual thresholds in the bass region translate to our non-anechoic listening rooms.

My understanding is that the ear cannot even detect the presence of bass energy from less than one full cycle. Considering the length of bass wavelengths relative to normal listening room dimensions, by the time the ear detects the presence of bass energy it has already begun arriving from multiple directions. So there is no perceptual (time-domain) separation between the first-arrival sound and the subsequent reflections in the bass region, and therefore the ear/brain system cannot determine the arrival direction(s).

My (sighted) observation has been that the more bass sources we have intelligently distributed around the room, the less localizable they each are, assuming none are distorting or passing audible upper bass/lower midrange energy due to inadequate lowpass filtering.

Then my questions are:
Is that true if the subs are in stereo? Is that true if the subs are in mono?
Does any variation in whether or not it is true depending on what frequency that it it crossed over at? And what about the crossover method and slope?
There seems to be more than one can of worms with this.

 

[Duke]

Then my questions are:
Is that true if the subs are in stereo? Is that true if the subs are in mono?
Does any variation in whether or not it is true depending on what frequency that it it crossed over at? And what about the crossover method and slope?

Multiple mono subs are less localizable than multiple stereo subs, assuming the bass energy being sent to the stereo subs actually is stereo.

The higher the lowpass frequency rolling off the top end of the subs, the more likely the sub(s) would be localizable.

The more gradual the slope of the lowpass filter, the more likely the sub(s) would be localizable.

There seems to be more than one can of worms with this.

The worms are much smaller and more benign than when you only have one subwoofer.

In practice with a distributed multisub system you can get away with a considerably higher crossover frequency than you could with just a single sub, unless that single sub is located somewhere between the main speakers.

My understanding is that the 80 Hz crossover frequency between sub(s) and main speakers that is more or less the standard for home theater was chosen as being adequate for when you have a single sub in a front corner. With a distributed multisub system you do not have that concentration of bass energy coming from just one side of the room, so the crossover frequency can be higher. I have set up distibuted multisub systems that worked well with a 150 Hz lowpass filter frequency, and I'm aware of at least one distributed multisub system that uses a 180 Hz lowpass filter frequency.

And if one of the subs IS localizable, with a distributed multisub system you have more freedom to re-locate that sub because the location of no single sub is critical. So if you can hear the location of the sub on the left-hand wall, you can move it further away from the listening location and probably solve that issue without having nearly as much effect on the in-room bass response as would be the case if you only had one subwoofer and you moved it.

 

[EJ3]

Multiple mono subs are less localizable than multiple stereo subs, assuming the bass energy being sent to the stereo subs actually is stereo.

The higher the lowpass frequency rolling off the top end of the subs, the more likely the sub(s) would be localizable.

The more gradual the slope of the lowpass filter, the more likely the sub(s) would be localizable.



The worms are much smaller and more benign than when you only have one subwoofer.

In practice with a distributed multisub system you can get away with a considerably higher crossover frequency than you could with just a single sub, unless that single sub is located somewhere between the main speakers.

My understanding is that the 80 Hz crossover frequency between sub(s) and main speakers that is more or less the standard for home theater was chosen as being adequate for when you have a single sub in a front corner. With a distributed multisub system you do not have that concentration of bass energy coming from just one side of the room, so the crossover frequency can be higher. I have set up distibuted multisub systems that worked well with a 150 Hz lowpass filter frequency, and I'm aware of at least one distributed multisub system that uses a 180 Hz lowpass filter frequency.

And if one of the subs IS localizable, with a distributed multisub system you have more freedom to re-locate that sub because the location of no single sub is critical. So if you can hear the location of the sub on the left-hand wall, you can move it further away from the listening location and probably solve that issue without having nearly as much effect on the in-room bass response as would be the case if you only had one subwoofer and you moved it.

I use 2 12" dual 4" voice coil (making each into a 4 Ohm circuit, ported cab (tuned to 29Hz) Floor Firing subs UNDER my mains with a Kilowatt of power each (from one bridged mono NAD 2200 amp on each one). 60 Hz low pass-80 Hz high pass.
It works ok & IF the signal is stereo that low, I can usually tell.

 

[prerich]

Why though? Or listening rooms are not anechoic chambers.

The author explicitly pointed out that the lack of room reflections in the test caused a significant change in the directional judgement.

So why would the results w/o room reflections be relevant to our listening environments w/ room reflections?

Here's another question - what if your speakers were designed to take the side reflections out of the equation, an open baffle speaker that creates a 90 degree null at their sides? this basically gets rid of first reflections.

 

[Cbdb2]

Here's another question - what if your speakers were designed to take the side reflections out of the equation, an open baffle speaker that creates a 90 degree null at their sides? this basically gets rid of first reflections.

No, it just shifts the first reflections to the front/back wall, floor or ceiling.