I am unclear on the two previous statements. How does the enclosure or driver Sd help with localization?
Since camplo insists that he can localize a 40hz tone produced to his subs' position it seemed to me that some secondary source would be the cause. A cavity resonance, one or more of the threaded rods he used to fasten the drivers (vibrating at a higher frequency) or an insufficiently supported enclosure panel. Any of which could show up with increased amplitude and is one of the reasons loudspeakers fail to disappear during playback in a room.
harmonic distortion makes a sub easier to localize.
If the woofer is firing at you perhaps is easier to localize versus down firing or slot loaded or push pull.
DJK said a sub should be between the mains.......... Or lower volume really close to you (due to the phase time delay just below the crossover point on either 12 or 18db crossover point, I can't remember).
On audiosciencereview, they lean towards a sub under each front, and / or quad spaced around the room subs.
If the woofer is firing at you perhaps is easier to localize versus down firing or slot loaded or push pull.
DJK said a sub should be between the mains.......... Or lower volume really close to you (due to the phase time delay just below the crossover point on either 12 or 18db crossover point, I can't remember).
On audiosciencereview, they lean towards a sub under each front, and / or quad spaced around the room subs.
At low volume the signal from 2 - 18"s should be free of noticeable distortion.
@gedlee how do we define "extreme neaefield"... I must be within the definition. I must be within the critical point....50/50 direct/indirect sound, from this point, increase direct sound = more locatable, increase room sound = less locatable.
This sounds realistic. Saying bass is unlocatable is too concrete of a statement.
Room Modes can hide the bass source location. Since room modes fluctuate and as well, proximity to the source is a factor, bass cannot always be unlocatable.
Bass being unlocatable in an anechoic chamber, are we saying the same result? I find it hard to believe... if you pointed one ear at the source and the other away, should not pressure be higher in the ear facing the source
Localization has to work In degrees meaning that "I can tell the general direction" vs "it sounds like it's coming from a very specific place in space" represent different levels of localization. To be "unlocatable" would say that it sounds like its coming from an area that it isn't or sounds like its coming from everywhere.
@gedlee how do we define "extreme neaefield"... I must be within the definition. I must be within the critical point....50/50 direct/indirect sound, from this point, increase direct sound = more locatable, increase room sound = less locatable.
This sounds realistic. Saying bass is unlocatable is too concrete of a statement.
Room Modes can hide the bass source location. Since room modes fluctuate and as well, proximity to the source is a factor, bass cannot always be unlocatable.
Bass being unlocatable in an anechoic chamber, are we saying the same result? I find it hard to believe... if you pointed one ear at the source and the other away, should not pressure be higher in the ear facing the source
Localization has to work In degrees meaning that "I can tell the general direction" vs "it sounds like it's coming from a very specific place in space" represent different levels of localization. To be "unlocatable" would say that it sounds like its coming from an area that it isn't or sounds like its coming from everywhere.
I think you are overthinking.
Well done subwoofer can not be localized. Period.
Do not try to play just the crappy sub and localize it.
Place two really good bookshelves in the right position, play them with and without the sub.
With the sub, it should be an improvement, since bookshelved do not go very low.
Can you localize sub in this situation? Than its not crossed low enough or it sucks! Period! Enough!
Well done subwoofer can not be localized. Period.
Do not try to play just the crappy sub and localize it.
Place two really good bookshelves in the right position, play them with and without the sub.
With the sub, it should be an improvement, since bookshelved do not go very low.
Can you localize sub in this situation? Than its not crossed low enough or it sucks! Period! Enough!
At 1meter.... if you can't localize at 1 meter, your ears are broken or the mode is so strong that pressure is everywhere instead of strongest coming from direction of sub...
Nothing crappy about this
Taken at 17"
The illusion of the room decimates with proximity, its that simple. Get close enough and yes, you can, localize, bass.
This is relevant as some people intend to listen this close, or at least, I do.
Nothing crappy about this
Taken at 17"
The illusion of the room decimates with proximity, its that simple. Get close enough and yes, you can, localize, bass.
This is relevant as some people intend to listen this close, or at least, I do.
Last edited:
It does work in degrees and the degree to which the low frequencies aid in the localization/image of a sound source is minimal. There are going to be instances in which you could localize a pure tone at LFs (such as in an anechoic chamber as you suggest or when very close to the source,) but these are extreme cases. When HFs are present they will overwhelm any localization cues from the LFs.
Sitting in a restaurant, small wide band speaker less than 2m from me, sub 8m away. Despite seeing the sub, it sounds like bass is coming from the small satellite. Although bass transients are a bit delayed
i'd say that the type of signal used can give localization cues... a transient like a kick drum where the fundamental does not time/phase align with the high frequency portion of the signal will sound like two sources rather then a cohesive single sound...our ear/brain extrapolates the error.
If you are trying to relate to my situation you are going to have lose about 7meters. I did state that if I move back farther, indirect sound takes over....
Quite true, though a properly designed full range system is hard to beat in terms of 'coherence'. Full horn systems excell at this.
Just thinking out loud, isn't the 100 ms onset still too short to prevent any HF content affecting the experiment? It would be interesting to start with no onset ramp and make it longer until it doesn't affect the results.
"In order to avoid any high-frequency content as a results of clicks and to
prevent the listener’s use of onset effects (spectral widening) as cues, all stimuli were
gated using 100ms onset and offset ramps."
"In order to avoid any high-frequency content as a results of clicks and to
prevent the listener’s use of onset effects (spectral widening) as cues, all stimuli were
gated using 100ms onset and offset ramps."
I would say no. 100 ms is sufficient to cover down to a few hundred Hz. But if there were transient clicks at the start of the ramping signal, they would be audible and lead to localization. I would hope that they minimized that issue. The ramp is necessary to minimize any transient HFs from the signal itself. They should be Gaussian pulses of tones or noise.
To me the experiment that I would like to see done would be to use broadband noise and move the sub. How many degrees would it take to resolve the displacement in that experiment? I would guess a lot if the movement could even ever be detected.
To me the experiment that I would like to see done would be to use broadband noise and move the sub. How many degrees would it take to resolve the displacement in that experiment? I would guess a lot if the movement could even ever be detected.
Still reading the paper but here is another on low frequencies:
https://www.researchgate.net/public...of_Human_Hearing_to_Changes_in_Phase_Spectrum
I was curious about the group delay of the hd650 headphones used in both papers.
https://www.researchgate.net/public...of_Human_Hearing_to_Changes_in_Phase_Spectrum
I was curious about the group delay of the hd650 headphones used in both papers.
Hi, unfortunately I do not have the GD matching that measurement. It would of been interesting to see how the GD turned out after voicing it that way. Obviously more like an anechoic measurement than the opposite, non the less, I was happy that I could achieve the response with not much effort. One of my first proofs of concept if you will.
This is the un-eq'd GD, theres room mode in this measurement though taken at about 1" in front of slot. The sim predicts 5ms at 30hz but the roll off slope is steeper in the sim as well. Also, a bit of delay is to be accounted for.