At lower frequencies room problems tend to be minimum phase and less location dependent. That's why PEQ works pretty good. When going higher up (around the Schroeder frequency) a lot more non-minimum phase behavior can be observed which is also very location-dependent. In that region it becomes very tricky to design a filter that doesn't do more harm than good.
No grey area with this one....!
Pinx,
With all due respect....How wrong can one be!
My points are not technical at all, they factually describe how perceive sound!!!!
The biology of how we perceive sound....I couldn't disagree with you any more clearly!
Cheers
Derek.
Pinx,
With all due respect....How wrong can one be!
My points are not technical at all, they factually describe how perceive sound!!!!
The biology of how we perceive sound....I couldn't disagree with you any more clearly!
Cheers
Derek.
In that case you know more than Jan Schnupp and everybody else in the vast field of auditory neuroscience. I'm impressed
Interesting example by how much phase can be distorted without much information loss:
Local time reversal and speech comprehension | Auditory Neuroscience
Local time reversal and speech comprehension | Auditory Neuroscience
I don't agree. Minimum phase means time invariant, not location invariant, i.e., the sound is not the same everywhere in the room, but remains the same over time at a given location. This is all explained in Toole's book. A 100 Hz wave is 11 feet long. Depending upon where you are sitting on the couch, you could be at the crest or the zero crossing of the wave. If you move, you are at a different point and a different correction is needed. At high frequencies, you can gate the response and correct the gated response. Basically, correct the direct sound at HF and direct plus reflected at LF. And with the latest tools, the transition between direct and direct plus reflected can be smooth, i.e., a continuously sliding time window.
Last edited:
Yes, the importance of phase is debatable. Of course it is great to have perfect phase and with the tools we have, this is very much doable. But is it audible? From the tests I did on myself, it was not audible to me. Now, if you use phase to change the magnitude response, of course it is audible. But if your magnitude response remains the same, and you flip the phase from standard (say with an LR4 crossover) versus flat, I was not able to detect the difference.
No MBL Omni for me!
Just to be clear I dont like Omni speakers, I dont use them or recommend them!
You have misunderstood my design goals completely, please see my point about recording technique below.
As it happens I believe the demand for large free standing high end loudspeakers (open baffle, Omni or whatever) has reduced to tiny niche demand only.
My market research tells me" make em tiny or make em on wall / in wall".
Thankfully one or two subs (sometimes 3 or 4) are acceptable domestically.
More importantly, re your point about about recording technique....I disagree here for sure!
This is a big point and is a very widespread misconception.....
One does not need to capture a 360 degree sound, to attempt to do so would be foolish.
Take a step back to the violin in the recording studio I posted a few pages ago.....In order for one to hear the violin is it a requirement:
(1) Our ears detect the full 360 degree sound "pattern"?
(2) Our ears detect any or all of the room reflections?
(3) We require and form of directivity manipulation?
Clearly the answer to all of the above is no.
If the violin was being played in an open field the 360 degree sound pattern (compression and rarefactions in air pressure) would spread out and diminish to infinity....Our ears would only detect the on-axis transient impulse response, therefore it follows that a microphone only needs to record the same information...!!!!
Another big misconception in the recording industry is the principal of multi mic'ing instruments such as drums / percussion / piano's etc. Even worse trying to capture the entire orchestra ....
Most recording engineers mix all the mic feeds into the same channel forgetting about the different arrival times of the sound (yes, good old compression and rarefactions) at the different mics....ie a close mic on the kick drum a few inches away or even inside the drum is mixed with a pair of flying mics 3 or 4 feet above the cymbals....
In conclusion....
The number one most important factor is time domain!
The goal should be to record time domain accurate impulses and then generate the closest possible replica of those transients with the broadest possible dispersion pattern in the vertical and horizontal planes.
Hope this helps and all the best
Derek.
Just to be clear I dont like Omni speakers, I dont use them or recommend them!
You have misunderstood my design goals completely, please see my point about recording technique below.
As it happens I believe the demand for large free standing high end loudspeakers (open baffle, Omni or whatever) has reduced to tiny niche demand only.
My market research tells me" make em tiny or make em on wall / in wall".
Thankfully one or two subs (sometimes 3 or 4) are acceptable domestically.
More importantly, re your point about about recording technique....I disagree here for sure!
This is a big point and is a very widespread misconception.....
One does not need to capture a 360 degree sound, to attempt to do so would be foolish.
Take a step back to the violin in the recording studio I posted a few pages ago.....In order for one to hear the violin is it a requirement:
(1) Our ears detect the full 360 degree sound "pattern"?
(2) Our ears detect any or all of the room reflections?
(3) We require and form of directivity manipulation?
Clearly the answer to all of the above is no.
If the violin was being played in an open field the 360 degree sound pattern (compression and rarefactions in air pressure) would spread out and diminish to infinity....Our ears would only detect the on-axis transient impulse response, therefore it follows that a microphone only needs to record the same information...!!!!
Another big misconception in the recording industry is the principal of multi mic'ing instruments such as drums / percussion / piano's etc. Even worse trying to capture the entire orchestra ....
Most recording engineers mix all the mic feeds into the same channel forgetting about the different arrival times of the sound (yes, good old compression and rarefactions) at the different mics....ie a close mic on the kick drum a few inches away or even inside the drum is mixed with a pair of flying mics 3 or 4 feet above the cymbals....
In conclusion....
The number one most important factor is time domain!
The goal should be to record time domain accurate impulses and then generate the closest possible replica of those transients with the broadest possible dispersion pattern in the vertical and horizontal planes.
Hope this helps and all the best
Derek.
ra7 said:
@ra7
- and because we can "separate out" sound coming from side angles it is "useless" to have influence e.g. on the levels of side reflections ?
(matching a speaker e.g. due to plain walls, wall carpets, furniture ... or to different positions and different toe in ...)
- and because whe can "separate out" those reflections they do not contribute to perceived quality and (thus) need no care/control ?
- and you cite Toole to back up this (your) point of view ?
Gosh, what a weired line of argumentation concerning side reflections IMO ...
ra7 said:
Let me put it like this:
It comes to a point (agreed!) when a desired amount of rise in DI should be realised by "narrowing" the radiation vertically more than horizontally.
Quite some (HF) units are known already, that take advantage of that to more or less extent. And such a beneficial "vertical to horizontal behaviour" is also implemented in my system, which i was mentioning (not describing) briefly.
Last edited:
@linearray:
It is easy to misunderstand someone when all you want to do is misunderstand.
I said the speaker directivity together with the room absorption is what we hear. I use absorption to control side wall reflections. But, in my opinion, one kind of directivity (narrow versus wide) is not necessarily better than the other. What Toole has shown is that side reflections can add a sense of spaciousness. In normal circumstances, sidewall reflections are never so loud that they are perceived as a second sound source. In normal circumstances, they almost always result in an addition of spaciousness.
I'm not going to argue with you anymore because all you are doing is trying to obfuscate matters rather than trying to further our understanding of the subject.
It is easy to misunderstand someone when all you want to do is misunderstand.
I said the speaker directivity together with the room absorption is what we hear. I use absorption to control side wall reflections. But, in my opinion, one kind of directivity (narrow versus wide) is not necessarily better than the other. What Toole has shown is that side reflections can add a sense of spaciousness. In normal circumstances, sidewall reflections are never so loud that they are perceived as a second sound source. In normal circumstances, they almost always result in an addition of spaciousness.
I'm not going to argue with you anymore because all you are doing is trying to obfuscate matters rather than trying to further our understanding of the subject.
Monaural phase errors are largely inaudible. However, binaural phase errors are highly audible. That's why I've said it's not productive to look at technical terms. They don't explain anything. They are good for describing something though. We need to look at perceptual properties.
Missing the point twice....!
Whilst I am flattered that you are impressed....
My warm & cuddly feeling is tempered by the fact that you have failed to understand the basics.....Quoting studies that explore more subtle aspects of a given subject does not compensate for your lack of understanding re the basics of that subject....Sorry!
Cheers
Derek.
In that case you know more than Jan Schnupp and everybody else in the vast field of auditory neuroscience. I'm impressed
Whilst I am flattered that you are impressed....
My warm & cuddly feeling is tempered by the fact that you have failed to understand the basics.....Quoting studies that explore more subtle aspects of a given subject does not compensate for your lack of understanding re the basics of that subject....Sorry!
Cheers
Derek.
Whilst I am flattered that you are impressed....
My warm & cuddly feeling is tempered by the fact that you have failed to understand the basics.....Quoting studies that explore more subtle aspects of a given subject does not compensate for your lack of understanding re the basics of that subject....Sorry!
Cheers
Derek.
Pardon my ignorance but if you feel I've missed the point twice the problem might not be the receiver but the transmitter? Care to elaborate on your findings in regard to audibility of phase distortion?
In the meantime others might find it interesting to read though this article:
Human Hearing - Phase Distortion Audibility Part 2 | Audioholics
Human Hearing - Phase Distortion Audibility Part 2 | Audioholics
ra7 said:
@ra7
I undestood very well and disagreed because of understanding very well:
The "common listener" does not like room treatment for many reasons. There are also "demanding" and "passionate" listeners, who do not like room treatment either:
A contradiction one might think, but a reality to me. There are not only "DIYers" and "the usual audio freaks" out there ...
ra7 said:
Well, it's more likely that i know some of my demanding customers (see above) and their habits better than you do ...
ra7 said:
I can see us agreeing in some "important core points", so no reason to be "nerved" on either side IMO ...
___________
Btw. i like the english verb "to obfuscate": That borrowing from latin sounds so strict and precise to my ears just in contrast to it's meaning.
Last edited:
Overkill,
I accept your clarification of what you were saying about capturing the sound field and not what I thought you were trying to say about an omnidirectional sound field. I am also in agreement that I just don't like omnidirectional speakers. Bose made enough mess of that back in the days of the 901, not something I care to have repeated.
Your statement about make them small is also something I agree on with today's consumer, at the same time what is considered small is a bit open to interpretation. I'm again not a fan of small 2" or 3" full range speakers with added subs, that just doesn't cut it to my ears. Okay for the typical passive listener I suppose on first listen though.
Glad to see you around and hope your project is coming along well. I did take your advice and have involved some others on the digital side of things, to much for one person to handle well and know all the best things to do.
On the subject of room placement I don't feel we have much control over that with the typical listener, they are going to put the speakers where they are most convenient or where they look best. So I say you better get the first arrival correct and that should be the dominant sound field, hopefully the room isn't so reflective it doesn't destroy the sound field but that is normally out of our hands.
Phase and impulse response at the crossover point is going to be the dominant factor with most two way or three way speakers, the extremes of highs and lows are not audible by the common listener, just my two cents on that part of the phase question.
That doesn't mean I want to ignore impedance rise or have massive phase wrap, just that I think what is critical is the combination of multiple devices at the overlaps.
I accept your clarification of what you were saying about capturing the sound field and not what I thought you were trying to say about an omnidirectional sound field. I am also in agreement that I just don't like omnidirectional speakers. Bose made enough mess of that back in the days of the 901, not something I care to have repeated.
Your statement about make them small is also something I agree on with today's consumer, at the same time what is considered small is a bit open to interpretation. I'm again not a fan of small 2" or 3" full range speakers with added subs, that just doesn't cut it to my ears. Okay for the typical passive listener I suppose on first listen though.
Glad to see you around and hope your project is coming along well. I did take your advice and have involved some others on the digital side of things, to much for one person to handle well and know all the best things to do.
On the subject of room placement I don't feel we have much control over that with the typical listener, they are going to put the speakers where they are most convenient or where they look best. So I say you better get the first arrival correct and that should be the dominant sound field, hopefully the room isn't so reflective it doesn't destroy the sound field but that is normally out of our hands.
Phase and impulse response at the crossover point is going to be the dominant factor with most two way or three way speakers, the extremes of highs and lows are not audible by the common listener, just my two cents on that part of the phase question.
That doesn't mean I want to ignore impedance rise or have massive phase wrap, just that I think what is critical is the combination of multiple devices at the overlaps.
@Kindhorman,
to my understanding neither Bose "901" nor e.g. Shahinian "Obelisk", "Hawk" are "omni" speakers.
IMO those designs (or contraptions) are at least aiming (among other things) at having even the reflections of lower order more diffuse (trying to use multiple reflections in front edges/corners of the room in case of "901") IMO and also having a reverberant field stronger than usual.
I would see the "Hawk" module somewhat "more successful" in going this way, whithout setting a value to that strategy in Loudspeaker Design itself from my side here.
From a pure "DI point of view" a Bose "901" is a rather highly directional speaker at least from midrange on upwards. DI itself does neither tell nor ask where to aim radiation lobes at.
Linkwitz "Pluto" e.g. is "quite omni" until a certain frequency range ...
Last edited:
LineArray,
then you can just jump to the MBL type if that is going to be your definition of omni. Doesn't mean I want that either. I just think you are introducing as many problems as you are trying to solve. Placement will be critical to say the least. Bose intent with the 901 was assuredly to create an omni directional sound field.
then you can just jump to the MBL type if that is going to be your definition of omni. Doesn't mean I want that either. I just think you are introducing as many problems as you are trying to solve. Placement will be critical to say the least. Bose intent with the 901 was assuredly to create an omni directional sound field.
The radiation pattern of the MBL Transducers is often referred to as being "radial":
mbl 101E Radialstrahler loudspeaker | Stereophile.com
http://www.stereophile.com/content/...-loudspeaker-measurements#p2JvpIRLDMlMYPmq.97
In this "101" model there are frequency regions, where the level is about 6dB down around +/- 15 degrees vertical already, also have a closer look at the top end >10Khz ...
If you want a prototypical "omni" i suggest using a Linkwitz "Pluto", it is "quite omni" up to 2-3Khz at least.
There are often great discrepancies between what many people think a certain LS design "should behave" (e.g. Bose 901, Shahinian "Obelisk", "Walsh Family", "MBL", "Linaeum", ...) and what those designs really show acoustically, especially at higher frequencies for the given drivers/transducers (due to polar pattern, DI) ...
A Bose 901 (again) can be expected to be "more directional" in mid-to-high range than a 2-way bookshelf having a small 4" woofer and and a 1" dome tweeter (without waveguide) and reasonably low crossover frequency say around 1,5 ... 2Khz.
Last edited:
Linearray,
I would say any speaker intending to radiate a 360 degree waveform in the horizontal axis around a center axis point would be considered an omnidirectional speaker no matter what its vertical dispersion happens to be. I know of no true omnidirectional speaker besides a corona discharge type of device, there are no other true pulsating spheres that I am aware of.
Bottom line is they don't recreate a true representation of a live event.
I would say any speaker intending to radiate a 360 degree waveform in the horizontal axis around a center axis point would be considered an omnidirectional speaker no matter what its vertical dispersion happens to be. I know of no true omnidirectional speaker besides a corona discharge type of device, there are no other true pulsating spheres that I am aware of.
Bottom line is they don't recreate a true representation of a live event.