Just a guess, but it looks like the horn is mismatched or too small. It could be digitally equalized, but I'd suspect directivity would be all over the place in the peak/dip region.
For example, one of the problems of classical exponential horns that are wider than they are tall is the notorious "pattern flip", which occurs one or two octaves above the published cutoff frequency. The crazy change in directivity, as well as the shift in overall diaphragm loading, is audible as a harshness, and some cases, grittiness that has given horns the "PA sound" reputation.
Although it might sound like I dismiss directivity, that isn't quite true. What I avoid are sharp changes in directivity with frequency (in less than an octave), or patterns that have sharp edges with ripples on the boundary. These correspond to changes in loading on the diaphragm, impedance variations, and reflections in the time domain.
Although the moderate T-ratio LeCleac'h horns are not constant directivity devices, they mimic the directionality and time response of direct-radiators, which is fine by me. I like direct-radiators and the way they sound and measure.
I just want more dynamics than you can get from soft-domes, which subjectively seem to compress the treble, at least compared to a good horn or AMT radiator. Soft domes are also grossly overtaxed if you try and take them down to 700 Hz, and mid domes have their own set of problems (diaphragm rocking due to lack of spider+surround suspension).
That's why I demurred earlier when asked about conventional soft-dome tweeters used in short waveguides. Just my feeling here, but that seems like the worst of both worlds: the potentially compromised time response of short horns, along with the modest dynamics of soft-dome tweeters.
The great virtue of the best soft-domes is outstanding time response; why goof it up by putting it in a short waveguide or horn? Getting good time response out of a horn or waveguide is non-trivial: it was reading Dr. Geddes' and Newell & Holland's books that underlined the point. You don't just slap it on, apply digital equalization, and hope for the best.
Last edited:
I'm not saying it can't be done: I just haven't seen any commercial examples where the waveguided soft-dome has better time-domain performance than a high-quality soft-dome on a flat baffle.
It's certainly possible to make a waveguide with good time response for a soft-dome driver, but I wouldn't expect it to be small, or for the waveguide to be operated through cutoff, which I what I've seen from commercial examples.
Commercial speakers, particularly those aimed at the home-theater market, typically use small (3~4" mouth) waveguides, which you'd expect to cut off around 2 kHz or so. That's pretty close to the crossover, and it has to be kept in mind that directivity control (and resistive diaphragm loading) disappears quite abruptly at cutoff.
It's certainly possible to make a waveguide with good time response for a soft-dome driver, but I wouldn't expect it to be small, or for the waveguide to be operated through cutoff, which I what I've seen from commercial examples.
Commercial speakers, particularly those aimed at the home-theater market, typically use small (3~4" mouth) waveguides, which you'd expect to cut off around 2 kHz or so. That's pretty close to the crossover, and it has to be kept in mind that directivity control (and resistive diaphragm loading) disappears quite abruptly at cutoff.
Last edited:
lots of thank for the answer, Lynn. I would like to use passive xo, if it's possible.
i'm trying to build a sort of econowave speaker. i have limited experience with speakers in general. my friend is a music lover, and have a Scan-Speak Event TL (18W/8535, D2905/9300), and after that he bought a pair of Final Sound 150i electrostatic panels. (he uses them with 8" Scan-Speak subs.) he (and I) prefers the sound of the panels compared to the TL.
but after that, I listened to 4Pi speakers with 2226h/de250 drivers, and I really liked it. (it took me a day to travel to listen them.) I really loved the effortless sound, dynamics and sound of conga, timpani. i can not go back to ordinary low-efficiency speakers anymore.
i'm trying to build a sort of econowave speaker. i have limited experience with speakers in general. my friend is a music lover, and have a Scan-Speak Event TL (18W/8535, D2905/9300), and after that he bought a pair of Final Sound 150i electrostatic panels. (he uses them with 8" Scan-Speak subs.) he (and I) prefers the sound of the panels compared to the TL.
but after that, I listened to 4Pi speakers with 2226h/de250 drivers, and I really liked it. (it took me a day to travel to listen them.) I really loved the effortless sound, dynamics and sound of conga, timpani. i can not go back to ordinary low-efficiency speakers anymore.
Last edited:
Lynn,
I did a study few months ago about the distortion of a 2" CD on a 300 Hz tractrix horn versus a bog standard cone/dome system, a B&W 602. The horn/CD combo sounds smooth, large and dynamic compared to the B&W, which sounds small, distant and distorted. The B&W actually has quite a smooth and flat response, but it just doesn't have the dynamics.
What I wanted to find out was whether distortion was causing the difference in dynamics. I was expecting to find the horn/CD combo would be clearly superior. But it wasn't. Here are the original results:
http://www.diyaudio.com/forums/mult...ns-beaten-behringer-what-120.html#post3394911
I subsequently measured the two speakers outdoors, and even the higher order distortion difference went away. So, why was I perceiving the difference in dynamics?
Tom Danley came up with a hypothesis that it is the direcitivity of the horn that makes it sound dynamic. He recommended taking the MTF and STIPA measurements in ARTA. He thinks they show the advantages of directivity.
http://www.diyaudio.com/forums/mult...ns-beaten-behringer-what-125.html#post3395529
Here are my measurements of MTF and STIPA of the B&W and 2445+tractrix combo:
http://www.diyaudio.com/forums/mult...ns-beaten-behringer-what-129.html#post3396323
Some differences begin to appear. If you follow the discussion in that thread and what Tom and Keyser point out, it is actually the direcitivity, and the increased clarity afforded by the higher direct to reflected ratio that causes us to perceive that immeasurable quality of dynamics.
A good example is to think of drums being struck in a highly reverberant room. It sounds loud. But if the drums are in a dead studio room, it doesn't sound as loud. It's the reflections and their level that make it seem loud.
And hence the next question, that if directivity is all we need, and if there isn't a real difference in performance in terms of distortion, then why not use the much smoother domes in a waveguide-like device?
Last edited:
Although I have great respect for Tom Danley, I cannot agree with the directivity hypothesis (tighter directivity results in impression of greater dynamics).
I've heard the LeCleac'h horns in several different contexts, and they sound even more dynamic than most horns, despite having a directivity pattern very much like a direct-radiator. What's unusual is the combination of horn dynamics with spatial qualities similar to an electrostat, or an MBL. Instead of tight and closed-in, the sound is wide-open and spacious, somewhat akin to the bigger Altec multicells (and nothing at all like the Altec sectoral horns).
My hypothesis is simply: greater headroom = greater impression of dynamics. A large-format compression driver has 10 to 20 dB more headroom than a 1" soft-dome, and this is directly audible, thanks to very brief transients not being compressed. Small-format compression drivers give up about 6 dB of headroom (compared to their big brothers), which is why I'm not as interested in them.
It has to be kept in mind that soft-domes are technically distorting all the time, due to deformation of the soft fabric of the dome, and greater accelerations result in more deformation of the dome. Direct-radiators are constant-acceleration devices, so more SPLs result in more G's on the diaphragm, which bends it more. Soft-domes have excellent self-damping, so there are no high-Q peaks (unlike metal diaphragms), but the deformation and lack of rigidity is there at all signal levels.
A similar thing goes on with midbass drivers. Polypropylene has excellent self-damping, but poly drivers much larger than 7 inches have a very dull, veiled sound. The only reason the Ariel gets away with poly is the drivers are very small. If you want to hear gruesome sound, try 12 and 15-inch polypropylene woofers. Paper is a good all-around compromise between self-damping and adequate rigidity, so it's been the large-format woofer cone of choice for the last eighty years.
Horns do not magically change the sound of the diaphragm. At best, they subjectively magnify the sound, providing more headroom, but they won't make a soft-dome sound like a metal dome, a metal dome sound like a soft-dome, or any other presto-chango magic. Phenolic diaphragms occupy in interesting middle ground, but their high mass makes them unsuitable for response extending to 10 kHz or beyond.
Beryllium, although costly at present, has a very interesting combination of rigidity and excellent self-damping. This diaphragm material works really well in direct-radiator dome tweeters (far superior to much more resonant titanium) as well as compression drivers. Now this is a dome tweeter that might indeed work very well in a medium-sized waveguide optimized for impulse response.
I've heard the LeCleac'h horns in several different contexts, and they sound even more dynamic than most horns, despite having a directivity pattern very much like a direct-radiator. What's unusual is the combination of horn dynamics with spatial qualities similar to an electrostat, or an MBL. Instead of tight and closed-in, the sound is wide-open and spacious, somewhat akin to the bigger Altec multicells (and nothing at all like the Altec sectoral horns).
My hypothesis is simply: greater headroom = greater impression of dynamics. A large-format compression driver has 10 to 20 dB more headroom than a 1" soft-dome, and this is directly audible, thanks to very brief transients not being compressed. Small-format compression drivers give up about 6 dB of headroom (compared to their big brothers), which is why I'm not as interested in them.
It has to be kept in mind that soft-domes are technically distorting all the time, due to deformation of the soft fabric of the dome, and greater accelerations result in more deformation of the dome. Direct-radiators are constant-acceleration devices, so more SPLs result in more G's on the diaphragm, which bends it more. Soft-domes have excellent self-damping, so there are no high-Q peaks (unlike metal diaphragms), but the deformation and lack of rigidity is there at all signal levels.
A similar thing goes on with midbass drivers. Polypropylene has excellent self-damping, but poly drivers much larger than 7 inches have a very dull, veiled sound. The only reason the Ariel gets away with poly is the drivers are very small. If you want to hear gruesome sound, try 12 and 15-inch polypropylene woofers. Paper is a good all-around compromise between self-damping and adequate rigidity, so it's been the large-format woofer cone of choice for the last eighty years.
Horns do not magically change the sound of the diaphragm. At best, they subjectively magnify the sound, providing more headroom, but they won't make a soft-dome sound like a metal dome, a metal dome sound like a soft-dome, or any other presto-chango magic. Phenolic diaphragms occupy in interesting middle ground, but their high mass makes them unsuitable for response extending to 10 kHz or beyond.
Beryllium, although costly at present, has a very interesting combination of rigidity and excellent self-damping. This diaphragm material works really well in direct-radiator dome tweeters (far superior to much more resonant titanium) as well as compression drivers. Now this is a dome tweeter that might indeed work very well in a medium-sized waveguide optimized for impulse response.
Last edited:
So, how would one measure headroom? I'm really asking. I want to find a measurement that correlates with the perception of dynamics. Distortion isn't it.
Are you talking about an impulse that a dome tweeter doesn't fully reproduce? Wouldn't that show up in a distortion test? Barleywater had done some burst cycle testing on tweeters. And basically, it showed that their distortion was almost an order lower than comparable compression drivers. See this post, and the explanation a few posts down:
http://www.diyaudio.com/forums/multi-way/221643-active-vrs-passive-80.html#post3243395
I am in complete agreement that large format CDs on the right horns sound magical. But just saying its headroom is not enough for me.
It's not just the directivity, it's the fact that you don't get early reflections. The direct sound is strong compared to the reflected level. That is what makes it clearer and us more tolerant of higher levels.
Are you talking about an impulse that a dome tweeter doesn't fully reproduce? Wouldn't that show up in a distortion test? Barleywater had done some burst cycle testing on tweeters. And basically, it showed that their distortion was almost an order lower than comparable compression drivers. See this post, and the explanation a few posts down:
http://www.diyaudio.com/forums/multi-way/221643-active-vrs-passive-80.html#post3243395
I am in complete agreement that large format CDs on the right horns sound magical. But just saying its headroom is not enough for me.
It's not just the directivity, it's the fact that you don't get early reflections. The direct sound is strong compared to the reflected level. That is what makes it clearer and us more tolerant of higher levels.
Last edited:
Lest not forget some of the newer composite diaphrams for CDs such as the Radians with an Al surface and Mylar surround. As a 1" goes, the 465 is one heck of a driver for the $$$.
As to waveguide mounted domes, I've experimented with quite a few and guides from MCM, Dayton and QSC. Of course a 1" exit horn is going to need modification to work with a dome but IME when the transition is correct, they sound wonderful.
.......and for dynamics, time domain accuracy and smooth response, true ribbons in their passband are hard to beat with the the larger AMT type units such as the TPL-150 really gaining ground as one of the best received HF drivers in a long time.
As to waveguide mounted domes, I've experimented with quite a few and guides from MCM, Dayton and QSC. Of course a 1" exit horn is going to need modification to work with a dome but IME when the transition is correct, they sound wonderful.
.......and for dynamics, time domain accuracy and smooth response, true ribbons in their passband are hard to beat with the the larger AMT type units such as the TPL-150 really gaining ground as one of the best received HF drivers in a long time.
A lower threashold of noise vs. signal with respect to room reflections. Yet not a lower threashold of noise vs. signal with respect to harmonic distortion?
In any event:
Most horns I've seen are setup where the integration window is far enough away from the listener that "not getting early reflections" is a myth (and brilliant marketing depending on the source).
Higher directivity and dynamics however.. to some extent yes, but probably little to do with room reflections. (..do you sound more dynamic singing in the shower (highly reflective environment), or singing in the closet (low reflective environment due to the hanging clothes)?)
Now higher directivity as an expression of compression and resulting mass.. maybe.
Last edited:
I agree with you that compression drivers distort at high output levels, I think largely because of their high compression. I have built basshorns with 1/3rd the distortion of compression drivers at the same (high) SPL, but they also have 1/3rd the compression ratio.
But compared to direct radiators, I have found that compression horns leave them in the dust in terms of SPL, so I think that's the main thing - their dynamic range. The direct radiating tweeters just don't have it.
At 100dB, most anything can get reasonable distortion figures, some even great. That's less than 10 watts for most direct radiators, and less than a watt for a horn. But between 110dB and 120dB, the scales begin to tilt. Some direct radiators can't reach those levels, and the ones that can need a lot of power to do it. The compression horns distortion rises, to be sure. It rises to double digits, even. But direct radiators often get into the triple digit range at these levels, and like I said, many can't even do it. They'll spit the dome out on the floor; Ribbons just go quiet.
So that's where I think the difference is, at levels higher than a relatively polite 100dB. I think it's 10-20dB above that where the horns show their strength. I guess that makes it really a matter of headroom to me.
But how often do peaks reach reach levels exceeding 100 db under most domestic conditions with music listening? Even some of the larger orchestral works rarely have moments where they do their +15 db crescendos. They are there (depending on the piece), but as an average the overall percentage has got to be low for most works.
Lot's of quantifiable specs to help with dynamics, there is no doubt about that, but it's those detail things that are not quantifiable that makes that final kill because it is impossible to measure and analyze everything to sufficient detail. This is the beauty of human intuition and intelligence. For example, with a same RC arrangement in a feedback and feed forward, I could change the sound quite significantly just by swapping sequence of the components where the analysis would say no change that can be considered audible. Then you get people to describe the difference, one could say there is a shift in tone balance, another can attribute to dynamics, yet another could talk about sound stage. They are all correct. But it is impossible to link the change in impression with one certain measurement. The real problem may not even be where you are looking.
Yes, generally 20 or 22dB below peak for classical recordings - sometimes more.
Getting that average level up to concert hall levels means the peaks are way up there. Horns do help, there.
My recent direct comparisons and frequency response measurements of LeCleac'h type 38" horn / LM 555, and Beyma 12p80nd brought quit surprising revelations. The 38"/LM555 has a rather wide dispersion pattern, and even more the Fane Studio 8m. But that doesnt help in subjective listening performance.
The 12p80nd as direct radiator sounds wide , open, with great clarity and transients, improves dramatically sound stage and depth, it sounds far more superior that the 38"/LM555/Fane Studio 8M in the lower midrange. Since i remember impressive performances of Avantgarde TRO, i suspect room size plays a important role. Maibe horns shine and show their strengths only in big rooms.
Certainly room interaction cannot be underestimated......
The 12p80nd as direct radiator sounds wide , open, with great clarity and transients, improves dramatically sound stage and depth, it sounds far more superior that the 38"/LM555/Fane Studio 8M in the lower midrange. Since i remember impressive performances of Avantgarde TRO, i suspect room size plays a important role. Maibe horns shine and show their strengths only in big rooms.
Certainly room interaction cannot be underestimated......
Hi Lynn, all
If subjective dynamics are the concern, exactly how could one go about measuring the dynamic performance of a loudspeakers?
At work, we use an independent lab to take some of the measurements and one of them does measure the maximum linear sound level. Here a musical spectrum pink noise signal is gradually increased every 5 min and at the point the response curve shape deviates 3 dB at any point from the 1w curve, that is the “max usable output”. Keep in mind, power compression with most modern drivers begins around 1/8 to 1/10 rated power. Naturally many go far beyond that level but at least up to that point, it is dynamically linear.
Another way to look at dynamics is “how fast” can the loudspeaker go from nothing to on to back off?
Here is why I am sure the MTF’s are at least holding the tiger’s tail and have proposed that connection.
In optics (I sort of a telescope guy at night) there is a common measure of resolution and it is called a Modulation Transfer Function. The idea is that it measures “how fast” the optic can go from black to white and in this case “how fast” is in special domain.
Google this later but now look at an example of an MTF in optics;
http://www.bobatkins.com/photography/technical/mtf/blur2a.gif
Here one can see that when the MTF is less than ideal, the blurring imposed gets worse the faster the rate of change. In a loudspeaker, picture he ideal loudness response is curve C while limiting the MTF produces a loudness response as in D. A and B show the dynamic effect.
For me, I had suspected a connection, the old Hopkins Stryker equation already pointed to directivity being a powerful factor for speech intelligibility but a seminar I went to this winter on predicting speech intelligibility (now a legal issue in Europe and to a degree here) made the light come on.
The legally binding STIpa prediction measurement is based on MTF measurements taken in 7 frequency bands and weighted to speech spectrum. The resulting data is crunched and ends with a STIpa number which is a language independent and proven accurate predictor of intelligibility (of random words say).
This is of great interest at work because many large buildings use the existing music line arrays for the emergency warning system and they often fail miserably when providing intelligible speech / warnings.
While MANY MANY things can reduce the MTF, ONE OF THEM is reflected sound / reverberant sound and here is where directivity can make a giant difference. In a soccer stadium in Germany, there was a comparison made between the 160+ line array boxes that were installed and 4 point source horns with high directivity. To hear the difference, they used half of it on one end and half on the other, the two point sources easily passed the intelligibility test, the company that installed the existing boxes said they needed 80 more and add absorption to reach passing. With the crowd and game noise, the announcer was unintelligible on one side and people laughed at his jokes on the other, with music, one side sounded like a concert, the other side sound like a hifi. This was an eyeopener.
Anyway MTF’s are an indicator of how well the source was able to follow the dynamics of the input signal. In playing around at home, I found that with any loudspeaker, the mtf got worse as you moved away from it (room stuff became louder compared to direct) AND there were pretty obvious differences when measuring at the LP depending on the directivity of the speaker. Anyway, ARTA has an MTF measurement, it is the rate of modulation (of the test frequency) and the achieved modulation depth.
What one sees is the higher the rate of modulation, the less modulation depth and this is the same as the blurring case in optics.
The real problem so far is that so much in audio is subjective and the connection between what one measures and hears is at times less than clear. So rather than say dynamics, what might be more accurate is to say the MTF’s are more like an indicator of preserving / transferring information to where ever the mic was.
Since hifi is subjective and based on single ended evaluation of recordings which if successful “sound like” a live image, it wasn’t until I found a non-subjective reference that I was sure. Here is a technique which uses intra aural crosstalk cancelation to make 3D stereo to work. This depends on delivering exactly the right signal to the opposite ear, it depends on the transmission of information accurately as possible and they concluded;
“Loudspeaker Directivity is the single most important loudspeaker property for 3D audio with crosstalk cancellation”
3D3A Lab at Princeton University
Not only that, but preserving the recorded stereo image also depends on the accurate transmission of information.
Lastly, we not only hear what the loudspeaker radiates but we can hear “how” the loudspeaker radiates spatially even if it shows nothing of interest in a measurement.
We hear in 3d, we measure in 1d. This part is something that appears to be very poorly understood yet is a key to making the loudspeaker “disappear” into a stereo image. These are not hifi speakers and not the most revealing BUT they do sound similar, have a very similar frequency response. One is a simple single point of radiation, the other is separate drivers.
Put on headphones and describe what you hear spatially in this video a fellow from Europe sent me.
https://www.youtube.com/watch?v=iJfH0pZve48
Anyway, got to run.
Best,
Tom
If subjective dynamics are the concern, exactly how could one go about measuring the dynamic performance of a loudspeakers?
At work, we use an independent lab to take some of the measurements and one of them does measure the maximum linear sound level. Here a musical spectrum pink noise signal is gradually increased every 5 min and at the point the response curve shape deviates 3 dB at any point from the 1w curve, that is the “max usable output”. Keep in mind, power compression with most modern drivers begins around 1/8 to 1/10 rated power. Naturally many go far beyond that level but at least up to that point, it is dynamically linear.
Another way to look at dynamics is “how fast” can the loudspeaker go from nothing to on to back off?
Here is why I am sure the MTF’s are at least holding the tiger’s tail and have proposed that connection.
In optics (I sort of a telescope guy at night) there is a common measure of resolution and it is called a Modulation Transfer Function. The idea is that it measures “how fast” the optic can go from black to white and in this case “how fast” is in special domain.
Google this later but now look at an example of an MTF in optics;
http://www.bobatkins.com/photography/technical/mtf/blur2a.gif
Here one can see that when the MTF is less than ideal, the blurring imposed gets worse the faster the rate of change. In a loudspeaker, picture he ideal loudness response is curve C while limiting the MTF produces a loudness response as in D. A and B show the dynamic effect.
For me, I had suspected a connection, the old Hopkins Stryker equation already pointed to directivity being a powerful factor for speech intelligibility but a seminar I went to this winter on predicting speech intelligibility (now a legal issue in Europe and to a degree here) made the light come on.
The legally binding STIpa prediction measurement is based on MTF measurements taken in 7 frequency bands and weighted to speech spectrum. The resulting data is crunched and ends with a STIpa number which is a language independent and proven accurate predictor of intelligibility (of random words say).
This is of great interest at work because many large buildings use the existing music line arrays for the emergency warning system and they often fail miserably when providing intelligible speech / warnings.
While MANY MANY things can reduce the MTF, ONE OF THEM is reflected sound / reverberant sound and here is where directivity can make a giant difference. In a soccer stadium in Germany, there was a comparison made between the 160+ line array boxes that were installed and 4 point source horns with high directivity. To hear the difference, they used half of it on one end and half on the other, the two point sources easily passed the intelligibility test, the company that installed the existing boxes said they needed 80 more and add absorption to reach passing. With the crowd and game noise, the announcer was unintelligible on one side and people laughed at his jokes on the other, with music, one side sounded like a concert, the other side sound like a hifi. This was an eyeopener.
Anyway MTF’s are an indicator of how well the source was able to follow the dynamics of the input signal. In playing around at home, I found that with any loudspeaker, the mtf got worse as you moved away from it (room stuff became louder compared to direct) AND there were pretty obvious differences when measuring at the LP depending on the directivity of the speaker. Anyway, ARTA has an MTF measurement, it is the rate of modulation (of the test frequency) and the achieved modulation depth.
What one sees is the higher the rate of modulation, the less modulation depth and this is the same as the blurring case in optics.
The real problem so far is that so much in audio is subjective and the connection between what one measures and hears is at times less than clear. So rather than say dynamics, what might be more accurate is to say the MTF’s are more like an indicator of preserving / transferring information to where ever the mic was.
Since hifi is subjective and based on single ended evaluation of recordings which if successful “sound like” a live image, it wasn’t until I found a non-subjective reference that I was sure. Here is a technique which uses intra aural crosstalk cancelation to make 3D stereo to work. This depends on delivering exactly the right signal to the opposite ear, it depends on the transmission of information accurately as possible and they concluded;
“Loudspeaker Directivity is the single most important loudspeaker property for 3D audio with crosstalk cancellation”
3D3A Lab at Princeton University
Not only that, but preserving the recorded stereo image also depends on the accurate transmission of information.
Lastly, we not only hear what the loudspeaker radiates but we can hear “how” the loudspeaker radiates spatially even if it shows nothing of interest in a measurement.
We hear in 3d, we measure in 1d. This part is something that appears to be very poorly understood yet is a key to making the loudspeaker “disappear” into a stereo image. These are not hifi speakers and not the most revealing BUT they do sound similar, have a very similar frequency response. One is a simple single point of radiation, the other is separate drivers.
Put on headphones and describe what you hear spatially in this video a fellow from Europe sent me.
https://www.youtube.com/watch?v=iJfH0pZve48
Anyway, got to run.
Best,
Tom
Hello,
Dynamics are not only related to the way the relative levels are reproduced.
IMHO a major origin of the impression that a good compression driver mounted on a good horn provides a better dynamics than other type of loudspeakers is linked to their better Impulse Response.
The wave emitted through a good horn from a good compression driver is mostly spherical in shape and the associated IR is very precise and thin like it is emitted by a point source. For the first arrival wave, a single path (a single ray) arrive to the ear of the listener.
With most others loudpeakers, they radiate the wave from a more complicate surface , the wavefront is more rough and also it is like the wave was emitted by a source having a complex shape, so rays with various pathlengths arrive to the ear. the associated IR is wider and less precise. For sure such case results in less resolution of the details of the music and IMHO this is a source of the loss of perceived dynamics obtained with conventionnal loudspeakers.
Best regards from Paris, France
Jean-Michel Le Cléac'h
Dynamics are not only related to the way the relative levels are reproduced.
IMHO a major origin of the impression that a good compression driver mounted on a good horn provides a better dynamics than other type of loudspeakers is linked to their better Impulse Response.
The wave emitted through a good horn from a good compression driver is mostly spherical in shape and the associated IR is very precise and thin like it is emitted by a point source. For the first arrival wave, a single path (a single ray) arrive to the ear of the listener.
With most others loudpeakers, they radiate the wave from a more complicate surface , the wavefront is more rough and also it is like the wave was emitted by a source having a complex shape, so rays with various pathlengths arrive to the ear. the associated IR is wider and less precise. For sure such case results in less resolution of the details of the music and IMHO this is a source of the loss of perceived dynamics obtained with conventionnal loudspeakers.
Best regards from Paris, France
Jean-Michel Le Cléac'h
Yes, but concert hall levels in a domestic environment? (..accepting a reduction in spl when factoring in much lower source distance.)
I doubt that for most people most of the time.
The real exception here are specifically those pieces of music where a fair bit of the average content is specifically played at very low "volume" - where the tendency is to increase the spl. Of course, the reverse is true as well - once that piece transitions to LOUD people tend to turn down the volume. I've seen it very often, and I've found myself doing the same. In fact, if we had "volume controls" at a live concert I'd bet everyone would be fiddling with the knobs.
A good horn system will only have the *objective* advantage under rather limited domestic conditions. (..and in truth, the real limitation isn't at higher freq.s for most systems, but rather at low freq.s. where you might be lucky to net 100 db at 30 Hz.)
As with much of audio, headphone listening presents a good counterpoint here.
Headphones have a very wide dynamic range throughout their pass-band, plus they have isolation a loudspeaker could only dream of - and yet, do they necessarily sound more "dynamic"? (..then there is the question of why do some headphones sound more "dynamic" than others with very similar design/construction and objective performance.)