ScottG, maybe we're using a different vocabulary but precise localization simply is not possible with multiple strong reflections present. That is exactly what happens when using omnidirectional speakers. They produce a very low direct/reflecting sound ratio. Download the reverberation example I posted – is the voice precisely localizeable for you?
markus76 said:
Fortunately most of the music I listen to isn´t produced that way. I have that problem with a few ancient music discs that unfortunately seem to be "produced" - awful word - by people who earn their living mainly with pop music. What interests me is just one thing - are the reverbs the problem or the people who use them?
Regards,
Oliver
Hi Earl.. yet again I'll differ.
It could well be that we aren't hearing the same thing, possibly because you have learned to listen to one set of features and I another?
As far as what I've suggesting being possible - of course its possible in even a modest room. It just requires moving your self and your speakers closer to the center of the room when you are ready to do some listening (move them back when you are done). In fact as I write this I've done just that. My speakers are about 6 ft from me (6 ft apart) and more than that distance from any wall (..though the one wall that can be "fudged on" is the one behind the listener). (..not coincidentally this isn't dissimilar "spacing" to a great number of mastering studio monitor placements.)
As for reviews, well there are any number of reviews that *generally* disagree. Not specifically in contrast to your own, but rather high freq. directive speakers (waveguide or not) that do not display phantom loci as well as other speakers.
It could well be that we aren't hearing the same thing, possibly because you have learned to listen to one set of features and I another?
As far as what I've suggesting being possible - of course its possible in even a modest room. It just requires moving your self and your speakers closer to the center of the room when you are ready to do some listening (move them back when you are done). In fact as I write this I've done just that. My speakers are about 6 ft from me (6 ft apart) and more than that distance from any wall (..though the one wall that can be "fudged on" is the one behind the listener). (..not coincidentally this isn't dissimilar "spacing" to a great number of mastering studio monitor placements.)
As for reviews, well there are any number of reviews that *generally* disagree. Not specifically in contrast to your own, but rather high freq. directive speakers (waveguide or not) that do not display phantom loci as well as other speakers.
None of the above - synthetic reverberation helps in translating a real acoustic situation or the idea of it to stereophony.
markus76 said:
You are wrong. Even with multiple strong reflections (assuming they aren't correlated in a negative fashion), precise "imaging" can be presented. DIRECT SOUND DOMINATES the presentation for localization (at higher freq.s in a typical domestic application).
Now strong reflections from the driver itself & the baffle.. THAT can screw up image localization.
When people have a problem with "omni's" (which really aren't omni's), it usually has to do with such surface reflections and the apparent width of acoustic center for a given freq.. Both can confuse "imaging" and "expand" it.
ScottG, aren't you aware that you ARE trying to minimize the impact of the indirect soundfield by choosing a small stereo triangle and increasing the distance of your speakers to room boundaries?
markus76 said:
Of course I am.. again, go back and read my first post.😉
My disagreements with some of your statements have been *particular*.
For instance:
"..precise localization simply is not possible with multiple strong reflections present. "
This is patently untrue in a typical domestic setting. Its not only possible, but rather it happens all the time even with a simply "good" audio system (regardless of radiation method).
or:
"..Razor sharp localizeable phantom sources are part of good imaging and are part of what stereophony is capable of. Omnidirectional speakers destroy that feature completely."
Again, incorrect. "Omni's" most certainly do not "destroy that feature completely" (..and I even provided *why* some "omni's" reduce this feature.)
Hello,
I just pulled the book out my bookself and found the picture you posted. You are right, it does propose the maximum spaciousness would be at 90 degrees.
But I cannot find any mention about the freq range this is valid. My copy is 1996, maybe you have a newer version and find the answer?
However I don't know how correct is this picture.
As I earlier mentioned Ando states 55 degrees.
See here:
http://www.ambiophonics.org/Ch_2_ambiophonics_2nd_edit.htm
it is said:
"The IACC depends mainly on the directions from which the early reflections arrive at the listener and on their amplitude. IACC measurements show a minimum at a sound source angle of 55 degrees to the median plane."
and it is continued:
"Ando points out that 90 degrees is not better because the almost identical paths around the head (front and back) double the leakage and, therefore, do not decrease the IACC effectively, particularly for frequencies higher than 500Hz."
So if interaural cross correlation IACC is a measure for spaciousness, who should we believe, Blauert or Ando?
Does anyone have the original Ando paper and would kindly check?
- Elias
markus76 said:
I just pulled the book out my bookself and found the picture you posted. You are right, it does propose the maximum spaciousness would be at 90 degrees.
But I cannot find any mention about the freq range this is valid. My copy is 1996, maybe you have a newer version and find the answer?
However I don't know how correct is this picture.
As I earlier mentioned Ando states 55 degrees.
See here:
http://www.ambiophonics.org/Ch_2_ambiophonics_2nd_edit.htm
it is said:
"The IACC depends mainly on the directions from which the early reflections arrive at the listener and on their amplitude. IACC measurements show a minimum at a sound source angle of 55 degrees to the median plane."
and it is continued:
"Ando points out that 90 degrees is not better because the almost identical paths around the head (front and back) double the leakage and, therefore, do not decrease the IACC effectively, particularly for frequencies higher than 500Hz."
So if interaural cross correlation IACC is a measure for spaciousness, who should we believe, Blauert or Ando?
Does anyone have the original Ando paper and would kindly check?
- Elias
Ando found different angles for different types of music. I'm sorry but I don't know what type of signal Barron used. He examined concert halls so I assume that he used music. Nevertheless it's uncertain what effect these reflections have on phantom sources. There simply are no scientific studies.
Best, Markus
Best, Markus
Scott - I think that you are knitpicking the discussion and not lending to it in a meaningful way. I completely agree with Markus even though his statements were a little more absolute than necessary. But then your statements have been absolute too and I find them to be "mostly" incorrect. What Marcus and I are saying is correct and based on very clear and well understood concepts in psychoacoustics.
For example your statement "DIRECT SOUND DOMINATES the presentation for localization ." is not really correct in the context of this discussion because the precidence effect, which I assume is what you are talking about, only implies the "principle" direction of the perceived image and does not apply to the clarity of the image. In other words the sound will appear to come from that direction dictated by the direct sound independent of any reflections, BUT the reflections will blur that image. This IS NOT good imaging in my world.
For example your statement "DIRECT SOUND DOMINATES the presentation for localization ." is not really correct in the context of this discussion because the precidence effect, which I assume is what you are talking about, only implies the "principle" direction of the perceived image and does not apply to the clarity of the image. In other words the sound will appear to come from that direction dictated by the direct sound independent of any reflections, BUT the reflections will blur that image. This IS NOT good imaging in my world.
markus76 said:
I discussed this topic at length with Heinrik Kuttruff some years back. He was in complete agrement with me that his work, that of Ando, etc. simply does not apply to a small room where the refections are spaced so closely to the direct sound. Toole claims to have data on the effect of very early reflections, but it is unpublished (until his book comes out). From what he told me I am not sure that I can completely agree with his conclusions (that early reflections DON't matter), but I need to see the data.
To get a handle on the very early refections you have to go to Blauert and gleen what you can from it since MOST of it is hard to apply not being specifically about the effects of early refections on image in two channels. Blauert is clear about image blur versus image location. To me simply creating the approximate location of a phantom source is NOT good imaging. It has to be precisely located with minimal blur. Early refections prevent this.
Hello,
Can you clarify the first sentence, do you suggest using omni below or above 800Hz?
What comes to omnidirectional monopole Linkwitz did explore it also in his Pluto, and at the time it was published it caused a lot of discussion. I think he also was talking about locating omnis close to you so direct to reverberant ratio is still big and imaging possible.
- Elias
ScottG said:
Can you clarify the first sentence, do you suggest using omni below or above 800Hz?
What comes to omnidirectional monopole Linkwitz did explore it also in his Pluto, and at the time it was published it caused a lot of discussion. I think he also was talking about locating omnis close to you so direct to reverberant ratio is still big and imaging possible.
- Elias
Once again, I'll disagree Earl.
You have both given absolute statements to support your position. If someone doesn't "speak up" who knows better, then its something of a disservice to others who read this thread.
You have chosen to specifically point out one area you found "mostly incorrect". And I've proven you wrong. Feel free to choose another statement. I'm not saying that I can prove you wrong again, but I'll give it a go.
.....
Your statement with regard to the precedence effect:
"..only implies the "principle" direction of the perceived image and does not apply to the clarity of the image. In other words the sound will appear to come from that direction dictated by the direct sound independent of any reflections, BUT the reflections will blur that image. This IS NOT good imaging in my world."
I'll not disagree with this, but:
Does this necessarily *negatively* effect localization?
and,
Isn't the "direction" in multiple dimensions a specific "3d" location?
also,
Does this necessarily effect to a *significant* degree "image" compaction or expansion?
IMO, it isn't a matter of IF the image is blurred (because it ALWAYS IS TO SOME DEGREE - and yes that's an absolute I'll stand by),
..but rather is the blurring harming the presentation? ..and (usually more often) if it is harming the presentation - to what degree is that harm in relation to the rest of the presentation? Moreover, can that harm be reduced without significantly harming other aspects of the presentation?
We may well have different definitions for what is "good imaging" and what isn't, but try thinking about it from my perspective instead of your own (..and yes, I've done the same). If and when you do so, please consider this:
"Imaging" is not *mastered* in a "vacuum" (..well, for most recordings). And in many instances it isn't recorded in such an environment either (..though vocals often are).
This means that usually (real or virtual) there are some ambient and "architectural" cues that provide "Hall Sound" in a recording (aka "spaciousness" and "soundstage").
These cues can provide a sense of space that the performers "play in".
Having these cues in addition to "imaging" cues tends to provide *better* localization of "imaging" because of a (recorded/same) reference. In other words your brain has more more information to properly place an image in context, and as a result its actually easier to localize and is less fatiguing to do so.
You have both given absolute statements to support your position. If someone doesn't "speak up" who knows better, then its something of a disservice to others who read this thread.
You have chosen to specifically point out one area you found "mostly incorrect". And I've proven you wrong. Feel free to choose another statement. I'm not saying that I can prove you wrong again, but I'll give it a go.
.....
Your statement with regard to the precedence effect:
"..only implies the "principle" direction of the perceived image and does not apply to the clarity of the image. In other words the sound will appear to come from that direction dictated by the direct sound independent of any reflections, BUT the reflections will blur that image. This IS NOT good imaging in my world."
I'll not disagree with this, but:
Does this necessarily *negatively* effect localization?
and,
Isn't the "direction" in multiple dimensions a specific "3d" location?
also,
Does this necessarily effect to a *significant* degree "image" compaction or expansion?
IMO, it isn't a matter of IF the image is blurred (because it ALWAYS IS TO SOME DEGREE - and yes that's an absolute I'll stand by),
..but rather is the blurring harming the presentation? ..and (usually more often) if it is harming the presentation - to what degree is that harm in relation to the rest of the presentation? Moreover, can that harm be reduced without significantly harming other aspects of the presentation?
We may well have different definitions for what is "good imaging" and what isn't, but try thinking about it from my perspective instead of your own (..and yes, I've done the same). If and when you do so, please consider this:
"Imaging" is not *mastered* in a "vacuum" (..well, for most recordings). And in many instances it isn't recorded in such an environment either (..though vocals often are).
This means that usually (real or virtual) there are some ambient and "architectural" cues that provide "Hall Sound" in a recording (aka "spaciousness" and "soundstage").
These cues can provide a sense of space that the performers "play in".
Having these cues in addition to "imaging" cues tends to provide *better* localization of "imaging" because of a (recorded/same) reference. In other words your brain has more more information to properly place an image in context, and as a result its actually easier to localize and is less fatiguing to do so.
gedlee said:
and yet it's true that such is His conclusion!
and it's Dr Toole not just "graaf"
isn't it shocking?
Hello Earl,
If you are talking about the audio acoustics ppt I remember reading that couple of years ago already. Also the pdf files I've seen.
You don't talk much about spatial hearing at high freqs. No mention about pinna localisation either. In the ppt you mention that image location is dominated in the range of 1-8kHz. That's about all. On the next page you already start talking about diffraction which concludes to your design of horns.
You have some useful information about distortion perception but that is not the one I'm looking for answers at this time.
May I ask if your goal in Summa has been to create 'they are here' or 'you are there' illusion?
- Elias
If you are talking about the audio acoustics ppt I remember reading that couple of years ago already. Also the pdf files I've seen.
You don't talk much about spatial hearing at high freqs. No mention about pinna localisation either. In the ppt you mention that image location is dominated in the range of 1-8kHz. That's about all. On the next page you already start talking about diffraction which concludes to your design of horns.
You have some useful information about distortion perception but that is not the one I'm looking for answers at this time.
May I ask if your goal in Summa has been to create 'they are here' or 'you are there' illusion?
- Elias
gedlee said:
Elias said:
Above 800 Hz. Particularly in the region of 1-4 kHz. Note that radially the Pluto almost achieves an "omni" dispersion in this range.
The two problems that people have with "omni's" and imaging are mostly related to those near boundary reflections (drivers, enclosures, and waveguides at higher freq.s), and apparent acoustic center for a given freq..
When some say an "omni" isn't "pin-point" enough they really aren't referring to image localization, but rather image *compaction*. However, others may have refined their hearing so that it actually is easier for them to *locate* an "image" that is more "compacted".
The best test is actually being present during a sound while its recorded (seated near the mic's during the recording). Playing it back, (though compared from memory), gives a good idea just how much or little image compaction is to reality. IMO "pin-point precision" is NOT realistic, nor is it necessarily better localized than something that isn't.
That is the single worst test one can do. A mic records a mono signal. This is completely different from how our hearing works. Your brain is capable of processing signals in time!
One mic records e.g. early reflections that produce comb filtering. That is clearly audible as heavy coloration when played back over speakers. Your ear won't hear any coloration at all when at the same location as the mic was.
Stop demanding stereophony being able to create a live like experience. It has strong limitations in that field but it has other features (razor sharp phantom sources is one of them) that make up the art of sound recording and mixing.
Best, Markus
Hello,
Yes I agree about low freq phantoms, with low I mean below about 700Hz or so. They are the only phantoms that can be reproduced in theory in stereo.
If this is so, the best is to design a highly directive speaker at LOW freqs, say below 1kHz.
In fact it may be better to have high freqs, say >1kHz, to have more wider dispersion than low end.
I'm proposing a concept totally opposite than conventionally seen.
- Elias
mige0 said:
Yes I agree about low freq phantoms, with low I mean below about 700Hz or so. They are the only phantoms that can be reproduced in theory in stereo.
If this is so, the best is to design a highly directive speaker at LOW freqs, say below 1kHz.
In fact it may be better to have high freqs, say >1kHz, to have more wider dispersion than low end.
I'm proposing a concept totally opposite than conventionally seen.
- Elias
- Status
- Not open for further replies.