Can the human ear really localize bass?

[gedlee]

It is over in Multi-Drivers or whatever it is called.

 

[bear]

markus,

comparison of the various methods showing non-flat SBA, I think this was the post.

Double Bass Array (DBA) - The modern bass concept! - Page 3

Here is a description - I know I saw a nice photo somewhere - of the variations including the 4 driver on the front wall arrangement.

As we all know, the simplest DBA is one speaker each in the center of the front and back walls.
The walls create acoustic images of these drivers every room-width to each side;
the ceiling and floor image the drivers every room-height above and below; and
this repeats, including all the diagonals, even out to infinity, for perfectly reflective walls, floors and ceilings.

A two-speaker DBA array centered vertically doubles the horizontal density of this simplest DBA,
but leaves the vertical density the same.
The two-speaker DBA array also shifts the driver horizontally by 1/4 of the room width, but the vertical offset is unchanged.
Nils's beautiful quad DBA then doubles the vertical density, too, also shifting the vertical placement by 1/4 room-height...

from this post: Double Bass Array (DBA) - The modern bass concept! - Page 4

Somewhere is the image of the quad of speakers on the front wall... too many pages...

 

[bear]

Oh! Shoot! It's the first post!!

ha ha.

http://www.avsforum.com/content/type/61/id/245169/width/500/height/1000/flags/LL

 

[markus76]

^
Bear, what are you trying to say? I don't follow.

 

[auplater]

Hmmnnn

I doubt it makes it unnecessary. It may be a pleasant effect, even with mono bass, but is it as desirable as uncorrelated bass that comes from the recording itself? (If the recording has it). In my listening I don't find that it is as satisfying or as real, but maybe some combination of the two could bring benefits.

I think you have a unique position on what "uncorrelated" means. In my world, uncorrelated means "not related to something else", so from that perspective, where the "uncorrelated bass" comes from is irrelevant, since it's not related to the source's origin

 

[Pano]

Well, yes - it's "not related" to the other channel. I.E., it's stereo. Of course the stereo channels are related, but each my contain unique materiel.

Not my term, it's Griesinger's. He was mostly talking about reverb being non-correlated between channels. We often do that in live sound to make a vocal or guitar sound bigger. It has to be adjusted for each venue.

If you mix the bass down to mono, you lose those differences between channels. No loss for some recordings, a definite loss for others.

 

[454Casull]

The DBA appears to be very appealing, but is likely only useful for mono bass.

_-_-bear

Absolutely not an issue if used for the LFE channel, as it sometimes (often?) is.

 

[gedlee]

What is being missed in the "correlated" channels discussion is that it is not a "is" or "isn't" thing. It a degree from uncorrelated to correlated (-1 > +1, actually, with zero being correlated). Most bass would be closer to 0 because even in live sound recording two separated mics are "highly correlated" at LFs. One could create a studio recording with completely uncorrelated bass, but I doubt that is very common or that many people would even know how to do it. So when discussing if the two stereo channels are correlated or not you have to define what you mean - are they correlated if only a small portion is correlated? Are they uncorrelated if only a small portion in uncorrelated. This whole discussion failed to quantify what was meant when these terms were used and as such any meaningful information was lost in the lack of common terminology.

 

[Pano]

And yet my tests showed significant differences below 100Hz in many recordings. Should those differences be correlated over a certain period of time to be considered stereo? I don't know. Should they be above a certain amplitude level to be considered stereo? Certainly - but what is that level? Still, the differences do exist on all the stereo jazz and classical recordings I tested.

We can go around and around about what's stereo and what's not, but in many recordings there is a significant difference between the 2 channels below 100Hz. The difference can be as much as in the midrange. Whether that's stereo or not, or if it matters to any particular listener are different questions.

We've been told again and again that human hearing can't locate bass, but with no good proof of that. And the received wisdom was that recordings are all mixed to mono in the bass anyway, which apparently is not true. It's up to you to decide if it matters to you or not.

 

[gedlee]

Pano the "best data" on your side is "it sounds that way to me". You know how far that goes to me.

The physics of the problem says that at some frequency localization must become impossible (that can be proven from the wavelengths and spacing of the ears - differences must vanish.) Is that frequency 50 Hz, or 100 Hz, or 10 Hz? Some reputable people have reported here that it is about 70 Hz. You report otherwise with tests of questionable scholarship.

I'm not going to argue with you because I have found that to be futile, but neither of us appears to have altered the others position one bit. I'm OK with that.

 

[bear]

Dr. Geddes,

One could argue that two LF sources at some distance from a listener and within some number of degrees of rotation off the center axis (to the front) of the listener will be indistinguishable as to source. Switch from speaker L to speaker R of axis.

Assume a mono signal for the moment.

Now, as the angle is increased, say all the way out to 90 degrees from the center axis left and right... driving one driver vs. the other driver, are we to expect that the source of that sound will not be detected??

Driving *both* drivers with the same signal and the same arrival times, it would be more difficult or impossible to properly detect the source (unsighted).

However this is where actual *stereo* bass comes in, most of the difference between channels *may* very well be small, but there is likely to be some phase/time difference and some freq resp difference. Clearly, if one flipped the phase of one of the two "90 degree" spread subs, one would not detect them as being "center"...

Does this make sense?

_-_-

 

[Pano]

Pano the "best data" on your side is "it sounds that way to me".

It is not, but that is what you continue to claim. That is quite false.

If I were selling and installing ultrasonic alarms that ran at 19Khz and telling everyone "It doesn't matter, no human can hear them", that would be just as false - and might be considered irresponsible. Some people can hear them are mightily annoyed by them. Probably you and I couldn't hear them, but that's no excuse for claiming that no one can hear them. It's simply false.

I do not have a vested interest in monophonic subwoofer systems, so there is no reason for me to claim that stereo bass is unnecessary. There is no vested reason for me to claim that stereo bass is better either - I don't sell subwoofers. I claim that for people who can hear stereo bass, it's more desirable then mono bass. I've confirmed that in listening tests, both blind and sighted. If you can't hear it, then it's of course no use to you. A color television might not be much use to a color blind man, either.

 

[dewardh]

This whole discussion failed to quantify what was meant when these terms were used and as such any meaningful information was lost in the lack of common terminology.

This whole discussion has failed to quantify even what is meant by "bass" or "subwoofer" . . . and much of the "difference of opinion" most likely derives simply from that. Not to mention that "which direction the buffalo are" is probably one thing out on the prairie, and another when they are "reproduced" in a 12ftx18ft room . . .

 

[Pano]

I thought there had been a general agreement in this thread that bass means frequencies below 100 Hz. Some people will allow up to 120, but 100 Hz seems a reasonable cut-off.
No agreement has been made on how steep a slope above the 100Hz mark. Personally I think 4th order is reasonable, but others may have tighter or looser views.

 

[daniel]

Everyone just add 'in the modal region' to stereo bass, no confusion then, right?

As a side note, there is a spatial audio committee in AES. Some years ago they found that in multichannel audio bass-managed x.1 is insufficient to generate spaciousness in the bass region. James Johnston and others suggested that full range x channel playback + >1 LFE sources is necessary to recreate believable spaciousness in the bass region. Can't locate the report, has anyone else seen something from them?
What that accomplishes is eerily similar to what Geddes suggests for multisub - multiple decorrelated low-frequency sources to get rid of the 'unnatural' small-room bass sound. That makes possible for the higher frequencies do the imaging and envelopment (requires multichannel) illusion without the bass constantly reminding you are sitting in a small room.

 

[auplater]

So I suppose the conclusion would be "IF you can locate the bass (in the modal region) in your current setup, your system is sub-par and you may want to reconfigure such that modal bass is decorrelated per AES, Geddes, et. als. to whatever extent necessary to provide greater realism and fidelity to the original venue.

Seems straightforward enough; I believe my current setup does just that, based on extended listening and some measurements.

But then, I have 8 drivers per primary (A & B), 8 drivers in surround, 4x12" subs spatially distributed, and one center channel, all driven through MultiEQ; total of 6x10" woofers, 8x4" midrange, 6 tweeters, and 2 BG RD 75" operating from ~500 Hz up. They all mesh seamlessly, and produce a quite convincing soundfield

You can see the A - B speakers + center in this photo:



John L.

 

[bear]

So, does this mean that in a small room that one *can* localize the bass, and this makes the room perception "small"?? The solution is to put the bass in an acoustic "blender" rendering the bass source "amorphous" and spread out - not localized?

Just trying to understand.

_-_-bear

John, that seems to be "rock" oriented set up there?

 

[dewardh]

I thought there had been a general agreement in this thread that bass means frequencies below 100 Hz.

Really? I got no such idea . . . where does that come from? If there's any "standard" for subwoofers it would be 80 Hz. 24dB (THX, chosen because that marks the lowest range of human voice), if there's any literal meaning to the word it's "below the woofers", which for any "full range" music speaker should be 30-40 to (at most) 50 Hz.

For vocalists "bass" has a different meaning altogether (and there's no doubt that we can all "localize" it), for the orchestra it's (probably) the octave below viola and bassoon generally occupied only by double bass, contra, tuba and piano (although it's not commonly played that low) . . . that would be roughly 32-64 Hz. and, as I've noted before, a range hard to localize in the concert hall, let alone the living room.

If your claim is only that you can localize 100 Hz. anechoic or in a room (warehouse) of more than three or four wavelengths shortest dimension then you won't find anywhere near as many skeptics as you will if the claim is that you can localize a 20 Hz. source in a 20x20 foot room. See what I mean by "undefined"?

 

[gedlee]

The issue here is smoothness and spaciousness in the bass versus a fixation on the traditional. You cannot have smooth response in the modal region with a single source (one channel of stereo if the bass is stereo). It is best if multiple sources are used and the locations of the mains is usually not a good location for one of these sources (although if there are enough it doesn't matter.) The multiple subs interact with the modes in a random fashion creating bass spaciousness. The use of a single LF source at a non-ideal location will not result in the best bass perception stereo or not. I would prefer the best bass perception over stereo bass and I would claim that most would as well.

But if you want to use multiple subs in stereo that won't hurt anything, I just don't think that it will be perceived any better than if the bass is mono at these frequencies.

In my setup there is stereo bass from the mains, remember I do not HP the mains, but the multiple subs are mono. To me this is the best practical setup.

The idea is de-correlated bass in the modal region, mono versus stereo is a distant second IMO.

 

[daniel]

"IF you can locate the bass (in the modal region) in your current setup, your system is sub-par and you may want to reconfigure such that modal bass is decorrelated per AES, Geddes, et. als. to whatever extent necessary to provide greater realism and fidelity to the original venue.

Excactly. The fullrange multichannel method (>=7! sources) lets to play with the amount of decorrelation, even if it is not worthwhile compared to multisub method. One thing is playing back 'stereo' sources (with all the bad recording practices), another is to recreate virtual environments (movies, games, simulators etc.)

So, does this mean that in a small room that one *can* localize the bass, and this makes the room perception "small"?? The solution is to put the bass in an acoustic "blender" rendering the bass source "amorphous" and spread out - not localized?

No, outdoors the bass is still correlated and localizable (marginally). The blender comes to play while simulating a different space that you are listening to.
What this accomplishes is simulating a modal-free region in the modal region - statistical region if you will. One could also argue that outdoors IS the ultimate big-space, modal density is infinite. There the modal region doesn't interfere with the localization of bass.
But localization is boring, mostly dominated by 700-3000Hz harmonics (onset of sounds). Envelopment and spaciousness is the interesting, less studied perception.