Interesting.
To the first I would say that it is incomplete in that as it stands it might be correct, but it ignores so much. How the DI is composed in time has to be a major factor in perception.
I am not sure that the second is necessarily true and certainly not once has a DI that contains the listeners.
Interesting, but I think fairly incomplete.
I have taken note of the fact that these are intended to be nearfield (I had missed that critical aspect). That is not a situation that I have ever dealt with, nor one that I particularly like. It's not all that common and certainly almost never used in a home setting. It does make a substantial difference in the design, but none of my experince is of much help in tis regard.
I work with nearfield placement since the early 80th. Yes it takes some time to get accustomed too. The advantage is not only that the speaker can be placed far from reflective surfaces ( except the floor ) but also that the speaker sounds louder with less input power required. A speaker that aproximates a point source loses 6dB per doubling of distance.
I think what TL is saying is pretty common thought in the pro monitor world. The most narrow monitor I've measured is marketed for midfield which I believe is generally considered 3-4 meters. THX PM3 compliant. Most people using these on the forums are still using them near-field for some reason and a few complain about their sweet spot being small compared to their other monitors.
Dan
Dan
Hi,
Not quite. But certain degree of preponderance of direct over reflected sound levels within a certain time window is needed to allow the spatial cues in the recording to be correctly rendered.
Why not instead make the directivity of the speaker adjustable?
As it stands we tend to design speakers for certain listening distances and rooms.
As it so happens, yes, but it requires adjustment for distance.
Realistically the acoustic behaviour of many rooms is quite similar, more similar than different certainly. This can and in my view should inform the design of speakers. Of course, some rooms differ dramatically and may require different speaker parameters.
Ciao T
Not quite. But certain degree of preponderance of direct over reflected sound levels within a certain time window is needed to allow the spatial cues in the recording to be correctly rendered.
Why not instead make the directivity of the speaker adjustable?
As it stands we tend to design speakers for certain listening distances and rooms.
As it so happens, yes, but it requires adjustment for distance.
Realistically the acoustic behaviour of many rooms is quite similar, more similar than different certainly. This can and in my view should inform the design of speakers. Of course, some rooms differ dramatically and may require different speaker parameters.
Ciao T
In my opinion the change in reflection patterns due to room shape and size, furniture, speaker and listener location should not be considered a minuscule factor in speaker design.
Hi,
Room shapes are actually often fairly similar. Furniture etc is hard to account for, but again, you are unlikely to find much between listener and speaker. So one can draw some general inferences.
But again, we need to consider the intended placement. My speakers are meant to be placed "conventional" in the mid- to far-field, this has implications regardless of furniture.
Equally, Joachim has a preferred placement where speakers and listener are near the middle of the room and close together but far from room boundaries (nearfield listening essentially). This too has implications.
Ciao T
Room shapes are actually often fairly similar. Furniture etc is hard to account for, but again, you are unlikely to find much between listener and speaker. So one can draw some general inferences.
But again, we need to consider the intended placement. My speakers are meant to be placed "conventional" in the mid- to far-field, this has implications regardless of furniture.
Equally, Joachim has a preferred placement where speakers and listener are near the middle of the room and close together but far from room boundaries (nearfield listening essentially). This too has implications.
Ciao T
After experimenting with tubes, cylinders, glass cups etc. i am back to the nylon sphere.
Looking at on and off axis responses of the midrange and tweeter that solution seems to do the least damage.
Looking at on and off axis responses of the midrange and tweeter that solution seems to do the least damage.
Preface for the censor.
The basis of the ZDL is a speaker that blends in and get lost in the listening space with little or no clue of its presence. Any sound radiated by the speaker that is not related to the program is a distraction and will alert the listener to its presence/location. Poor driver size selection for the required sound pressure output may increase the effects of diffraction and also generate other sucking, wheezing or otherwise easily locatable sounds.
Hello ,
I went to the symphony last night and came away with a couple of thoughts (dangerous? Yes!).
The orchestra will not fit in my listening room. Related to this thought is creating the illusion of them being there is quite a challenge (#1).
I have never seen a speaker driver that looks like a cello. Related to this thought the mechanisms these two items use to generate sound are worlds apart. Getting them to sound the same to my ear-brain is another challenge (#2).
#1 has to do with recording methods, speaker placement and other stuff. Let this one sit on the table.
#2 To keep the model simple think a full range driver .Take your red felt tip pen and draw a dot on the driver. To produce a realistic sound pressure level that driver needs to move a quantity of air. To move that quantity of air that driver of a given cross sectional area needs to move at a specific velocity over a specific distance. Watch that red dot how fast does it accelerate and how far does it move. How much like a cello is it? That acceleration, velocity and distance strikes me as being quite different than a cello. I am thinking that the driver needs to match the frequency spectrum and minimize the distance traveled, piston velocity and acceleration to match more closely the sound of a cello. I am thinking slew rate kind of stuff, Doppler effects and (fill in any items of your own.) The motions of the driver that are unlike the cello may produce distracting sounds that subtract from the illusion of being there.
DT
The basis of the ZDL is a speaker that blends in and get lost in the listening space with little or no clue of its presence. Any sound radiated by the speaker that is not related to the program is a distraction and will alert the listener to its presence/location. Poor driver size selection for the required sound pressure output may increase the effects of diffraction and also generate other sucking, wheezing or otherwise easily locatable sounds.
Hello ,
I went to the symphony last night and came away with a couple of thoughts (dangerous? Yes!).
The orchestra will not fit in my listening room. Related to this thought is creating the illusion of them being there is quite a challenge (#1).
I have never seen a speaker driver that looks like a cello. Related to this thought the mechanisms these two items use to generate sound are worlds apart. Getting them to sound the same to my ear-brain is another challenge (#2).
#1 has to do with recording methods, speaker placement and other stuff. Let this one sit on the table.
#2 To keep the model simple think a full range driver .Take your red felt tip pen and draw a dot on the driver. To produce a realistic sound pressure level that driver needs to move a quantity of air. To move that quantity of air that driver of a given cross sectional area needs to move at a specific velocity over a specific distance. Watch that red dot how fast does it accelerate and how far does it move. How much like a cello is it? That acceleration, velocity and distance strikes me as being quite different than a cello. I am thinking that the driver needs to match the frequency spectrum and minimize the distance traveled, piston velocity and acceleration to match more closely the sound of a cello. I am thinking slew rate kind of stuff, Doppler effects and (fill in any items of your own.) The motions of the driver that are unlike the cello may produce distracting sounds that subtract from the illusion of being there.
DT
It will be intersting to see John´s solution in the midrange. It is there where i am strugeling the most. Say i use the up facing isobaric to 250Hz with an LR4. Then we have
- 6dB at 250 Hz and can asume that the woofer works as an omni. Then in the midrange we place the second crossover at 2.5kHz. The midrange will be omni at deeper freqencies and the go more and more directional at higher frequencies. The tweeter at the lower end of the passband will start as an omni too. I think i have to define over what angle i can keep the energy flat. After the measurements and tests i made i asume it could be done resonable well over an angle of 120°.
What is your opinion about frequency response ? Should i make it flat on axis or should i modify the response so that the speakers listenen too from an angle sounds flat ? My usual setup is that the speakers are 2.85 meters apart ( distance tweeter-tweeter ) and i listen at 2 meters or less from the speakers ( distance tweeter-ear ).
- 6dB at 250 Hz and can asume that the woofer works as an omni. Then in the midrange we place the second crossover at 2.5kHz. The midrange will be omni at deeper freqencies and the go more and more directional at higher frequencies. The tweeter at the lower end of the passband will start as an omni too. I think i have to define over what angle i can keep the energy flat. After the measurements and tests i made i asume it could be done resonable well over an angle of 120°.
What is your opinion about frequency response ? Should i make it flat on axis or should i modify the response so that the speakers listenen too from an angle sounds flat ? My usual setup is that the speakers are 2.85 meters apart ( distance tweeter-tweeter ) and i listen at 2 meters or less from the speakers ( distance tweeter-ear ).
Last edited:
Hello Joachim,
It sounds like your on/off axis system performance and measurement is going according to plan. It is not typical of what is seen down at the high end store. I think that you are authoring the book. A little trial and error to see what you like?
My thought would be an extended sweet spot in the near field love seat for two. The whole room would be a reach.
That chair, is that the Edison electric chair created during the Power Wars between Edison and Tesla?
DT
All just for fun!
It sounds like your on/off axis system performance and measurement is going according to plan. It is not typical of what is seen down at the high end store. I think that you are authoring the book. A little trial and error to see what you like?
My thought would be an extended sweet spot in the near field love seat for two. The whole room would be a reach.
That chair, is that the Edison electric chair created during the Power Wars between Edison and Tesla?
DT
All just for fun!
Two people could listen well if we cross the speakers before the listeners and take atvantage of the time of flight - visa - intensity tradeoff. The OX tweeter has a rising response anyway so we can afford to listen slightly off axis. Yes, DT, someday i will make it work.
DT, Mr.Stoll has done it successfull for decades so i do not claim any authorship. The only thing i have done is to put some of his ideas into public.
DT, Mr.Stoll has done it successfull for decades so i do not claim any authorship. The only thing i have done is to put some of his ideas into public.
I can't see the pictures, so I'n not sure. But numerous configurations have been considered over the years, just none that seemed practical enough to fit in a normal living environment. It's never been a high priority item because of that.
Basically, the chair has to look stylish, sound has to be excellent in the seat, but ambient sound with just the seat playing will be also good. Surround sound, options to install gaming gadgets that fold out of sight easily, also act as a media control chair.
Last edited:
Hi,
Let's not, because here it becomes very crucial.
I would like to propose two very contrasting scenarios...
I call them the "I am there" and "They are here" scenarios. Both can be applied to listening with minimalist recordings, that preserve the spatial cues of the original auditorium, massive multi tracked recordings using closeup microphones generally do not apply.
The "they are here" refers to a feeling that the musicians are in the room. This can be great for a small Jazz ensemble, but works poorly for even quite small classical ensembles, in most rooms there is not enough space, sure the whole room is filled with sound, but any pretensions of realism are lost.
The "I am there" refers to a feeling that one has been transported out of the familiar surroundings of home into a concert halls, Walls disappear, orchestras assume realistic proportions etc...
Both of these extremes have specific requirements.
I find that the "they are here" type of presentations needs very wide dispersion and ton's of reflections, one might say a low ratio of direct to reflected sound.
To remove the walls and to be transported into a larger acoustic space requires that reflections are minimised.
If we now consider the two kinds of diffraction, one which is simply the "bending of waves around obstacles" and the other, which is the "creation of directional, nonflat frequency response sound radiation from sharp corners", we find that "bending around obstacles is a problem for "I am there" but helps for "They are here", while the "radiation from sharp corners" is undesirable for both.
However, as the "I am there" reaction requires strong directivity, the presences of sharp corners generally is no issue, while the wide dispersion for "They are here" really does not tolerate them well.
I have never seen a microphone that looked like a cello either, or a camera that looked like an actress.
To some degree the recording aparatus is an abstraction of the way our senses work, the camera is an abstracted eye (not a very good one) and the microphone is an abstracted ear (not a very good abstract).
The reproduction side generally is the recording device in reverse, e.g. film projector is a camera in reverse, speaker is a microphone in reverse.
These high levels of abstraction do have their challenges when it comes to providing convincing illusions. For example, as exciting as 3D Movies are (especially in an iMax cinema), I am not fooled for a second.
It seems easier to suspend disbelief with eyes closed and a good audio setup + good recording.
This argues for the use of very large drivers, so movement is minimised.
What I have found is that very large, very directive speaker systems (e.g. Tannoy 15" Coaxials) correctly set up are exceptional at producing the "I am there" illusion, in ways smaller systems cannot. I have never heard a system with a 5" midrange that could manage that.
In some ways even cheap old 10"-12" Full Range drivers (if not forced to overexert themselves with producing bass) are doing great in that department, despite all their other faults and lack of absolute fidelity.
What I would find interesting is to have Joachim's view on how his setup with a closeup position of the speaker and wide dispersion and small drivers compares and why.
What I can see is that close distance means small speaker drivers are not as stressed. So they may have a similar lack of distortion and great subjective dynamics as large far field systems (though my personal experience with small nearfield systems does NOT support this contention, small nearfield systems still sound small and dynamically challenged, even close up).
The close distance means a good direct to reflected ratio, despite a wide dispersion speaker. In this the system may be similar to large, highly directive far field systems. One interesting property is that low level long time window reverb will be much more dense and higher in level for the small wide dispersion nearfield system than for the large, highly directive far field system.
I am unsure how this would effect things, with full range dipoles (ESL's especially) the long delayed extra reverb caused by the rear radiation seems to create a feeling of spaciousness at the expense of image specificity.
If the angle subtended by the speakers is equal to that of a narrow dispersion, far-field, large system we MAY (and this is one of my questions, as my experience with near field systems is that this is precisely NOT the case) acquire a similar sonic perspective.
Ciao T
Let's not, because here it becomes very crucial.
I would like to propose two very contrasting scenarios...
I call them the "I am there" and "They are here" scenarios. Both can be applied to listening with minimalist recordings, that preserve the spatial cues of the original auditorium, massive multi tracked recordings using closeup microphones generally do not apply.
The "they are here" refers to a feeling that the musicians are in the room. This can be great for a small Jazz ensemble, but works poorly for even quite small classical ensembles, in most rooms there is not enough space, sure the whole room is filled with sound, but any pretensions of realism are lost.
The "I am there" refers to a feeling that one has been transported out of the familiar surroundings of home into a concert halls, Walls disappear, orchestras assume realistic proportions etc...
Both of these extremes have specific requirements.
I find that the "they are here" type of presentations needs very wide dispersion and ton's of reflections, one might say a low ratio of direct to reflected sound.
To remove the walls and to be transported into a larger acoustic space requires that reflections are minimised.
If we now consider the two kinds of diffraction, one which is simply the "bending of waves around obstacles" and the other, which is the "creation of directional, nonflat frequency response sound radiation from sharp corners", we find that "bending around obstacles is a problem for "I am there" but helps for "They are here", while the "radiation from sharp corners" is undesirable for both.
However, as the "I am there" reaction requires strong directivity, the presences of sharp corners generally is no issue, while the wide dispersion for "They are here" really does not tolerate them well.
I have never seen a microphone that looked like a cello either, or a camera that looked like an actress.
To some degree the recording aparatus is an abstraction of the way our senses work, the camera is an abstracted eye (not a very good one) and the microphone is an abstracted ear (not a very good abstract).
The reproduction side generally is the recording device in reverse, e.g. film projector is a camera in reverse, speaker is a microphone in reverse.
These high levels of abstraction do have their challenges when it comes to providing convincing illusions. For example, as exciting as 3D Movies are (especially in an iMax cinema), I am not fooled for a second.
It seems easier to suspend disbelief with eyes closed and a good audio setup + good recording.
This argues for the use of very large drivers, so movement is minimised.
What I have found is that very large, very directive speaker systems (e.g. Tannoy 15" Coaxials) correctly set up are exceptional at producing the "I am there" illusion, in ways smaller systems cannot. I have never heard a system with a 5" midrange that could manage that.
In some ways even cheap old 10"-12" Full Range drivers (if not forced to overexert themselves with producing bass) are doing great in that department, despite all their other faults and lack of absolute fidelity.
What I would find interesting is to have Joachim's view on how his setup with a closeup position of the speaker and wide dispersion and small drivers compares and why.
What I can see is that close distance means small speaker drivers are not as stressed. So they may have a similar lack of distortion and great subjective dynamics as large far field systems (though my personal experience with small nearfield systems does NOT support this contention, small nearfield systems still sound small and dynamically challenged, even close up).
The close distance means a good direct to reflected ratio, despite a wide dispersion speaker. In this the system may be similar to large, highly directive far field systems. One interesting property is that low level long time window reverb will be much more dense and higher in level for the small wide dispersion nearfield system than for the large, highly directive far field system.
I am unsure how this would effect things, with full range dipoles (ESL's especially) the long delayed extra reverb caused by the rear radiation seems to create a feeling of spaciousness at the expense of image specificity.
If the angle subtended by the speakers is equal to that of a narrow dispersion, far-field, large system we MAY (and this is one of my questions, as my experience with near field systems is that this is precisely NOT the case) acquire a similar sonic perspective.
Ciao T
Joachim,
Yes. With a small box and 250Hz crossover you can even put the woofer on the back and it will be omni.
Yes, the tweeter to midrange crossover becomes the problem (always).
The tweeter can be solved.
One way is to look at the 45 or 60 degree off axis curves of your midrange and treble and to set the crossover point so that there is a smooth handoff off axis, if the drivers are flat on axis the result on axis will be good too. If you need to equalise the drivers you need to look at the equalised off axis response.
As I personally always try to make narrowish dispersion speakers I use the 30 Degree off axis response after EQ as my main determinant. No matter how, I find doing it that way promotes low crossover points.
The other solution is to apply some form of absorptive (felt mask) or reflective (horn/waveguide) form of directivity control to the tweeter, so that the crossover region has no directivity step. Or just use a bigger tweeter, such as the 34mm Audax, which then raises the necessity of a supra-tweeter, as larger tweeters both get very directional and loose output at higher frequencies.
The final solution is to go 4-way, with a cone midrange in a sphere or tapered tube combined with a 2" Dome midrange (crossover at around 1.2KHz) also in a sphere or tapered tube and a 20mm Dome tweeter in a sphere or tapered tube on top, with maybe around 5.1KHz Crossover between upper midrange and treble.This brings us to something that is quite B&W Nautilus-ish in concept.
Despite their other drawbacks, it would seem to me that coaxial may answer the purpose more well than you are willing to grant and promote simplicity.
I have two separate issues here.
1) As you will be in the nearfield, I think you need to account for the so-called baffle step, or baffle loss, in your case probably in the midrange. In speakers designed for farfield use this is normally a "bad idea" (which does not stop loads of people applying it).
2) Unless you can incorporate asymmetric directivity patterns (like in the old JBL Everest) over a wide bandwidth I think it is best to aim for constant flatish, gently falling response across a wide angle.
Ciao T
Yes. With a small box and 250Hz crossover you can even put the woofer on the back and it will be omni.
Yes, the tweeter to midrange crossover becomes the problem (always).
The tweeter can be solved.
One way is to look at the 45 or 60 degree off axis curves of your midrange and treble and to set the crossover point so that there is a smooth handoff off axis, if the drivers are flat on axis the result on axis will be good too. If you need to equalise the drivers you need to look at the equalised off axis response.
As I personally always try to make narrowish dispersion speakers I use the 30 Degree off axis response after EQ as my main determinant. No matter how, I find doing it that way promotes low crossover points.
The other solution is to apply some form of absorptive (felt mask) or reflective (horn/waveguide) form of directivity control to the tweeter, so that the crossover region has no directivity step. Or just use a bigger tweeter, such as the 34mm Audax, which then raises the necessity of a supra-tweeter, as larger tweeters both get very directional and loose output at higher frequencies.
The final solution is to go 4-way, with a cone midrange in a sphere or tapered tube combined with a 2" Dome midrange (crossover at around 1.2KHz) also in a sphere or tapered tube and a 20mm Dome tweeter in a sphere or tapered tube on top, with maybe around 5.1KHz Crossover between upper midrange and treble.This brings us to something that is quite B&W Nautilus-ish in concept.
Despite their other drawbacks, it would seem to me that coaxial may answer the purpose more well than you are willing to grant and promote simplicity.
I have two separate issues here.
1) As you will be in the nearfield, I think you need to account for the so-called baffle step, or baffle loss, in your case probably in the midrange. In speakers designed for farfield use this is normally a "bad idea" (which does not stop loads of people applying it).
2) Unless you can incorporate asymmetric directivity patterns (like in the old JBL Everest) over a wide bandwidth I think it is best to aim for constant flatish, gently falling response across a wide angle.
Ciao T
- Status
- Not open for further replies.