The whole d*mn family
Ugly cousin
Very rich cousin
My favorite cousin
See Dave, vibrating toroid segment, schmegment.😀
I'll call 'em cousins if I want to. 😉
Cheers,
AJ
An externally hosted image should be here but it was not working when we last tested it.
Ugly cousin
An externally hosted image should be here but it was not working when we last tested it.
Very rich cousin
An externally hosted image should be here but it was not working when we last tested it.
My favorite cousin
An externally hosted image should be here but it was not working when we last tested it.
See Dave, vibrating toroid segment, schmegment.😀
I'll call 'em cousins if I want to. 😉
Cheers,
AJ
Madmike2 said:
More or less all of them... speaker building is a balanced set of compromises...
dave
paulspencer said:
Beaming is a loss of spl on both sides (horizontally) of the reference axis (which is usually intended to be "flat" in response). (Non-traditionally however it can also mean a speaker that has a "rising response" with a sp loss off axis (physical axis), typically where the off-axis response is intended to be the reference axis.) Graphically* you'll usually see a tweeter with a flat on-axis (the reference) response to 20 kHz (and the reference axis and physical axis are the same). At 15, 30, and 45 degrees off of that axis (on both sides horizontally) you'll have a progressive loss in spl (i.e. at 15 degrees at 20 kHz the response is down 6 db, at 30 degrees the response is down 12 db, at 45 degrees the response is down 20..etc.) To a degree this is beaming, however when someone usese the term "beaming" they are usually stating that the loss is more severe off-axis AND that it usually begins lower in freq. than normal, (i.e. where a normal tweeter may be "flat" at 3 kHz off-axis at 15, 30 and even 45 degrees, a beaming tweeter may be down several db's at 15 degrees - and more as we go further off-axis horizontally).
*graphically the best examples I've seen are the latteral response graphs that JA produces in Stereophile when measuring a speaker.
Beaming is directly related to the width of the radiator (tweeter), the wider it is the worse the beaming will be. This is why when you look at the graphs of tweeters (and other drivers) the smaller the diameter of the driver typically the higher in freq. it will go with out an appreciable loss of sp off-axis. Electrostatic panels that are flat and wide (and are intended to run full range) often "shoot" for a flat on-axis response and will have an off-axis response that starts droping in spl as low as 800 Hz.
Beaming can be usefull in some situations but is almost never a good thing (at least "traditional" beaming is almost never a good thing). In some situations it can be used (as Thorsten aka KYW alluded to) to achieve a specific purpose (..like INCREASING the spl off-axis to compensate for a listener moving off-axis - here the speaker has a non-flat "rising response" which means that the reference axis is actually physically off-axis, i.e. the speaker is meant to be listened to off-axis, in this instance then the speaker is NOT traditionally beaming).
The most often proposed purpose (of directionality in general) is to reduce room effects. Of course reducing room effects to an extent is a laudible goal, but doing so in a loudspeaker's design CAN be a matter of creating a bigger problem than the one you are fixing - and this is particularly true with respect to higher freq.s.
Room effects contain two basic problems Grouping and Reflections.
Grouping is the effect where the rooms nodes, (in the modal region - which is dependent on the room's size, geometry, and construction) start to create deviations in freq. response (typically around 200 Hz and below - to a point ..dependent on the room). Often the deviations are at least as much as + or - 6db. There is no real solution to this problem (without creating other problems), but the current DIY fad today is dipole radiation which usually gets a pretty even freq. response for the listener over a fairly broad coverage area. In any event, beaming is irrelevant to grouping because beaming is concerned with freq.s higher than 200 Hz.
Reflections can take the form of correlative and non-correlative reflections.
Correlative reflections are those where at least one reflection is within the wavelength's length.. i.e. the length of a 3 kHz signal is a little over 4 and 1/2 inches - IF the transducer (tweeter) reproducing this freq. is this close to a reflection point then there is a high degree of correlation. The higher in freq. this occurs (up to about 13kHz), usually the worse it is - and can cause all sorts of audible problems, it however is usually not a significant problem with respect to room reflections simply because the rooms boundries (particularly horizontal boundries i.e. walls) are not close enough to be problematic.
Non-correlative reflections are of course the opposite. They ARE a problem, but the question is: how big of a problem? The common response is that they are a significant problem, however this is usually untrue. Consider that direct sound is dominantly percieved by the listener over reflected sound even if reflected sound is considerably higher in spl than the direct sound (..provided that there is a low degree of correlation). Because of our ability to "spot" direct sound "over" reflected sound, non-correlative reflections are usually a minor subjective problem. The subjective description of this problem is often described as the "hall of mirrors" problem. Usually imaging will be detrimently impacted subjectivly with a sense of "blurring" to images (or diffuse images). The degree of this blurring is usually increased as the reflections are closer in time to the direct sound and as they are higher in spl. These non-correlative reflections do not effect image position to any great extent. Interestingly enough, the more uniform the dispersion is in spl (i.e. an "omni" is maximally uniform), the more uniform the blurring is - and is virtually always preferred subjectivly for a given spl.
Another problem that room reflections can have is with overall spl percieved by the listener - this is a freq. balance thing where a speaker that beams can be problematic (..or "good" thing depending on the recording). If the recording is "HOT" where the freq. balance is effectivly non-flat and "tilted upward", then beaming can be good because the reduction in off-axis sound higher in freq. by the beaming speaker produces a "downward tilt"at the listener's position (which effectivly cancels-out the "upward tilt" of the recording). (Note though that this is an oversimplification and the only consistently effective solution is a non-detrimental eq. that can be adjusted by the user at their listening position for each recorded track.) If the recording is NOT "HOT" and the speaker beams then the overall subjective response by the listener is usually that the system is "bass heavy" with a "downward tilting response". Of course the LEVEL of the non-correlative reflections are the key ingredient here - the higher in spl's they are with respect to higher freq.s, the greater the spl at those freq.s at the listener's position.
Konnichiwa,
I merely pointed out that I am familiar with the products from Waves. And the EQ Plugins I have at least are not FIR (and should not be at any extent).
Are you going to stop reading stuff into my writing that is not there?
I did not promote any ill defined narrow dispersion, but CONTROLLED dispersion with a precise set of parameters as to HOW controlled.
And Dome tweeters remain unsitable for the job unless utilised with a horn. This is most basic acoustics.
What CD Spinner?
Why do I need a £ 900 DAC? Explain. And I hope you realise that I have more than one PC around, including one for media center and music (Foobar at that), Crossovers as such are not needed, neither is with the current setup room EQ/DRC.
I am keeping an open mind.
But facts remain just that.
You may wish to apply yourself to the material on recording and acoustics posted in my technical discussion group. It MAY make some of my points clearer.
But here it is again:
If you wish to replay stereo recordings made for loudspeaker reproduction with a resonable degree of accuracy a number of conditions MUST be fulfilled by the combination of listening space and replay system. One place where these are well summarised are the IRT Monitoring Recommendations (I'm happy to e-mail you a copy).
Failure to fulfill these conditions causes deviations from the original. These may or may not be desired by the individual listener.
Take it or leave it.
Sayonara
ShinOBIWAN said:
I merely pointed out that I am familiar with the products from Waves. And the EQ Plugins I have at least are not FIR (and should not be at any extent).
ShinOBIWAN said:
Are you going to stop reading stuff into my writing that is not there?
I did not promote any ill defined narrow dispersion, but CONTROLLED dispersion with a precise set of parameters as to HOW controlled.
And Dome tweeters remain unsitable for the job unless utilised with a horn. This is most basic acoustics.
ShinOBIWAN said:
What CD Spinner?
ShinOBIWAN said:
Why do I need a £ 900 DAC? Explain. And I hope you realise that I have more than one PC around, including one for media center and music (Foobar at that), Crossovers as such are not needed, neither is with the current setup room EQ/DRC.
ShinOBIWAN said:
I am keeping an open mind.
But facts remain just that.
You may wish to apply yourself to the material on recording and acoustics posted in my technical discussion group. It MAY make some of my points clearer.
But here it is again:
If you wish to replay stereo recordings made for loudspeaker reproduction with a resonable degree of accuracy a number of conditions MUST be fulfilled by the combination of listening space and replay system. One place where these are well summarised are the IRT Monitoring Recommendations (I'm happy to e-mail you a copy).
Failure to fulfill these conditions causes deviations from the original. These may or may not be desired by the individual listener.
Take it or leave it.
Sayonara
I know newbies should stay out of stuff like this, but reading this thread and seeing references to Linkwitz and his tweeter choice I was reminded of something he said about dispersion:
"I have observed that a certain amount of lateral reflection is necessary for creating a believable illusion of the space in which the recording took place or to give a satisfyingly rich musical experience [3,4,5]. Flush (soffit) mounted speakers strike me as producing a sound that is two-dimensional, analytical, but ultimately lifeless. Highly directional horn speakers create a similar negative impression. In the tweeter range, wide horizontal dispersion removes harshness and adds airy-ness in my experience. This only makes sense if the reflected sound intensity has been increased. It might also explain why ribbon tweeters, which can be very narrow sources leading to wide dispersion, are so highly regarded, despite their poor vertical polar response."
Sorry to jump in with no thoughts of my own, I'm just beginning the journey.
"I have observed that a certain amount of lateral reflection is necessary for creating a believable illusion of the space in which the recording took place or to give a satisfyingly rich musical experience [3,4,5]. Flush (soffit) mounted speakers strike me as producing a sound that is two-dimensional, analytical, but ultimately lifeless. Highly directional horn speakers create a similar negative impression. In the tweeter range, wide horizontal dispersion removes harshness and adds airy-ness in my experience. This only makes sense if the reflected sound intensity has been increased. It might also explain why ribbon tweeters, which can be very narrow sources leading to wide dispersion, are so highly regarded, despite their poor vertical polar response."
Sorry to jump in with no thoughts of my own, I'm just beginning the journey.
Konnichiwa,
I am familiar with Mr. Linkwitz's view.
I am also familiar with making minimally miked recordings of acoutical music and monitoring the using highly directional soffit mounted speakers. What the monitors tell you of the Mike Feed is what you hear when you place your head about where the Mikes are. If that results in a recording not "airy" enought of sounding too "two-dimensional" then the microphone arrangement must be changed and perhaps also the damping of the acoustic space.
We now are at the point where we have to accept that many recordings are flawed. In which case it is perfectly acceptable to compensate for the problems using effects, however such effects ideally have a "bypass" switch.
For fun, I'm working conceptually on a speaker that will offer fully steerable directivity seperately for LF and MF/HF allowing to be dialed through omnipole, dipole and cardiode continously (not really difficult, it has been done for microphones for ages).
Such a system then allows the direcivity of the Speaker to be matched to the recording. If you want accuracy adjust for hypercardiode response, if you want to turn a very sparse and minicuse recorded sound image into a room filling one select omnipole and so on. Tuning LF and MF/HF directivity separatly allows to keep a minimal room mode excitation while freeing up the Mid/Treble unit for any particuar desirable radiation pattern.
Given the minimal interest from any side in such a speaker development will take a long time though and will probably only be possible if other projects provide the profits to do so.
Anyway, some notes from here....
Sayonara
poptart said:
I am familiar with Mr. Linkwitz's view.
I am also familiar with making minimally miked recordings of acoutical music and monitoring the using highly directional soffit mounted speakers. What the monitors tell you of the Mike Feed is what you hear when you place your head about where the Mikes are. If that results in a recording not "airy" enought of sounding too "two-dimensional" then the microphone arrangement must be changed and perhaps also the damping of the acoustic space.
We now are at the point where we have to accept that many recordings are flawed. In which case it is perfectly acceptable to compensate for the problems using effects, however such effects ideally have a "bypass" switch.
For fun, I'm working conceptually on a speaker that will offer fully steerable directivity seperately for LF and MF/HF allowing to be dialed through omnipole, dipole and cardiode continously (not really difficult, it has been done for microphones for ages).
Such a system then allows the direcivity of the Speaker to be matched to the recording. If you want accuracy adjust for hypercardiode response, if you want to turn a very sparse and minicuse recorded sound image into a room filling one select omnipole and so on. Tuning LF and MF/HF directivity separatly allows to keep a minimal room mode excitation while freeing up the Mid/Treble unit for any particuar desirable radiation pattern.
Given the minimal interest from any side in such a speaker development will take a long time though and will probably only be possible if other projects provide the profits to do so.
Anyway, some notes from here....
Sayonara
Jeez, I'd rather use "home-speak"
ScottG just posted the most thorough, helpful response to the question. IMHO. There is another side of the issue. An approach from the backside will broaden the perspective and (hopefully) allow each participant to make the best decision for themselves.
Search for "waveguides", and/or "Dr. Earl Geddes". In spite of having been pounced upon already, an understanding of horns and waveguides can make the decision making more fruitful.
There are many tools available to us in the box. We find ourselves using those tools in grossly different venues. The search for the right tool is a lifelong pursuit. The decision can occur in a moment. It is elegant.
Another consideration: Wilson's Maxx and JMLabs Utopia series (to name some) set "m's" on their baffles angled with respect to each other. Does this mitigate beaming? I think so.
ScottG just posted the most thorough, helpful response to the question. IMHO. There is another side of the issue. An approach from the backside will broaden the perspective and (hopefully) allow each participant to make the best decision for themselves.
Search for "waveguides", and/or "Dr. Earl Geddes". In spite of having been pounced upon already, an understanding of horns and waveguides can make the decision making more fruitful.
There are many tools available to us in the box. We find ourselves using those tools in grossly different venues. The search for the right tool is a lifelong pursuit. The decision can occur in a moment. It is elegant.
Another consideration: Wilson's Maxx and JMLabs Utopia series (to name some) set "m's" on their baffles angled with respect to each other. Does this mitigate beaming? I think so.
AJ,
that Vifa tweeter without the phase plug looks very interesting. If I were looking for a dome tweeter it would be high up on my list. Eff is very good. The SS looks nice but I'd be too stressed about damaging such an expensive tweeter. I've just set up my speakers active with no protection. I think it will be ok but if something goes wrong its not much of a hassle to replace it. The SS retails for almost as much as I could build my existing mains for!
So why did you choose the Vifa XT19? Are you aiming to take a wide dispersion tweeter and add some directivity with a waveguide? ... and hence match the polar response of the Seas 8"?
Another q ... how do you find that magnesium driver? I have been inclined to avoid them as being a bit revealing (I hate bad recordings but seem to collect them as if it were a hobby!)
that Vifa tweeter without the phase plug looks very interesting. If I were looking for a dome tweeter it would be high up on my list. Eff is very good. The SS looks nice but I'd be too stressed about damaging such an expensive tweeter. I've just set up my speakers active with no protection. I think it will be ok but if something goes wrong its not much of a hassle to replace it. The SS retails for almost as much as I could build my existing mains for!
So why did you choose the Vifa XT19? Are you aiming to take a wide dispersion tweeter and add some directivity with a waveguide? ... and hence match the polar response of the Seas 8"?
Another q ... how do you find that magnesium driver? I have been inclined to avoid them as being a bit revealing (I hate bad recordings but seem to collect them as if it were a hobby!)
If you find the time, investigate, its great stuff.
What do you defined as controlled. The ATC mid has a shallow horn/waveguide as does the SS ring, these are controlled directivity. But just which parameter do you look for which deems a driver to have a certain threshold of acceptable controlled dispertion which makes DRC/EQ and room treatment optional?
Simple answer is you cannot answer that question, the room is the problem not the speaker and every room is different.
Forgive my ignorance but I cannot for the life of me remember what that quasi-tube based SACD player is called. I know the name has on oriental flavour. Its in your pic with your wife and speakers in the center of your kit.
I think you missed my point, I was trying to highlight how ones preference for audio solutions can come across as forced and all knowing.
But explaining why you need a £900 clock (I mistakenly called this a DAC on my previous post) is very much like you being in preference to forced dispertion. I see is as a quality boost in much the same as you see yours. What I don't do is slam other methods which I see as potentially inferior, or different, down.
I could also go onto mention that whilst foobar offers lossless bitperfect playback, the soundcard, clock, DAC's and interelated components are very important, irrelevant of EQ/DRC or and other sound shaping tools. So you may not be experiencing the best that a PC solution has to offer. You could list your equipment, I could pick holes with it, but its diversity that makes things interesting and equally valid.
There is no one way to achieve perfect audio, our understanding of psycho acoustics in theoretically simulation is fine for perfect case scenario's such as rooms with no furniture or precisely modelled absorbant factor of a particular carpet at a certain frequency. But the a living room is anything but perfect case. Your forced directivity is no more valid a fix than my DRC/EQ and room treatments. Yet again its another way to acheive the same, the end results are largely based on the proficiency of the implementations and room your dealing with.
Here's my room and the placement of speakers, listening postion and treatments.
An externally hosted image should be here but it was not working when we last tested it.
Black denotes Auralex room treatment and grey is the seating area.
I've used a very extreme case of 90 degree radiation pattern with assumed distributed loudness jthroughout this pattern(which is actually a good thing) just to show a worse case scenario for room reflections. You'll consider that there are actually no early horiizontal reflections with my room and Auralex DST and wedges are used for HF absorbtion on late reflections, First relfections off the floor aren't taken care aside from carpet. The SS ring rad doesn't actually have anything like a 90 radiation pattern but for arguments sake.
Bass is very difficult to pratically tame with traps, especially the stuff below 80hz. So EQ and placement all contribute to acheiveing a flat response for me.
I think its best to consider each case on its merits and not encapsulate forced dispertion as a way to defer on EQ/DRC and room treatments. Likewise DRC/EQ and room treatments aren't the holy grail but can help in many situations, including those relating to forced dispertion systems, when applied correctly and sensibly.
My room is wide with the speakers well away from the walls so forced dispertion is less valid. Each method is equally valid and offers no increase in subjective sound quality if the room is moved out of the equation. Therefore you should decide what is best for you and your room and remember that a wide dispertion system with treatments can sound better than a forced dispersion without and the opposite is equally true. Also remember that forced dispertion doesn't mean minimal relfections the room decides that and your certainly not immune from late reflection from the rear of the room.
Finally remember that dispertion is a small part of the whole and tweaking for dispertion doesn't mean great sound. Though I agree in multiway system you should certainly design for it for smooth response over a large sweet spot and I did. Just not to the extent that it null's any improvements DRC/EQ/Treatments offer - that's just not possible.
planet10 said:
Agreed.
Its the ones that claim there's only one way to do things that you need to take with a pinch of salt.
Konnichiwa,
The basic principle is primitive.
If I combine a figure 8 (dipole) Mike and omni Mike (omnipole/monopole) I can mix the amplitudes of both at different levels and if neccesary add small time delays and as a result obtain any directivity from omnidirectional via cardiode & hypercardiode to dipolar. The only requirement is that our mike array must be smaller than a significant fraction of the wavelength of the signal we wish to capture.
A speaker can be made to steer directiity on the same principle, obviously the ability to precisely steer directivity becomes problematic as wavelength become very small and there are other issues attached, including that of in effect requiring a fully active system with many different individual amplifiers which at least among audiophiles is considered badly chosen, as their favourite amplifiers cannot be used and issues around requiring a pair of quite non-directional treble radiators (a problem in itself).
I do believe I have answers to most of these issues, of course the whole thing comes 20 odd years too late for the market and for me too, why did I not know and understand 20 Years ago what I do now... ;-)
Sayonara
paulspencer said:
The basic principle is primitive.
If I combine a figure 8 (dipole) Mike and omni Mike (omnipole/monopole) I can mix the amplitudes of both at different levels and if neccesary add small time delays and as a result obtain any directivity from omnidirectional via cardiode & hypercardiode to dipolar. The only requirement is that our mike array must be smaller than a significant fraction of the wavelength of the signal we wish to capture.
A speaker can be made to steer directiity on the same principle, obviously the ability to precisely steer directivity becomes problematic as wavelength become very small and there are other issues attached, including that of in effect requiring a fully active system with many different individual amplifiers which at least among audiophiles is considered badly chosen, as their favourite amplifiers cannot be used and issues around requiring a pair of quite non-directional treble radiators (a problem in itself).
I do believe I have answers to most of these issues, of course the whole thing comes 20 odd years too late for the market and for me too, why did I not know and understand 20 Years ago what I do now... ;-)
Sayonara
Konnichiwa,
Simple. A directivity that remains constant with frequency up to around 3 - 5KHz and goes down ideally to DC and which remains either constant or preferably smoothly and slightly narrows above 3 - 5Khz, so that reflected sound follows the "house curve" (as it should, unless you want more "air" than a concert hall provides).
I suggested a DI of 6db up to the 3 - 5KHz boundary and then an increase smoothly over two octaves to a DI of 10db, in other words around 90 degree spherical -6db beamwidth below 3 - 5Khz and around 60 degrees spherical -6db beamwidth at the top of the audio band.
You may find this to be actually comparably wider at high frequencies than many common tweeters manage yet comparably narrower at lower frequencies than most tweeters will control and equally narrower than what most woofers/midranges will control through the reverbant range of the room.
They are controlled directivity for part of their usable bandwidth, in either case the waveguide is by far too small to show any effect near the lower operating frequency limit (simply look at waveguide diameter and wavelength).
They are NOT controlled diretcivity and in many ways worse than no controll at all.
(Additional) Room treatment becomes optional if the RT60 falls around 0.1 - 0.3 Seconds above 300Hz and if the Speaker has a DI of at least 6db but no more than 10db below 3 - 5KHz.
DRC is optional if the sound output from the speaker system is reasonably flat over a range of around 100Hz-10KHz (the IRT prefers 60Hz - 12.5KHz, better safe then sorry).
Both items are achievable by combining normal designs of rooms with suitably directional speakers.
With my "commercial" hat on I modify Shanling Gear, so I often have it at home. My "own" Digital players are two.
One is a heavily modded 2nd generation Pioneer DVD Player which actually plays CD and DVD Disks asyncronous (DVD Disks are always anyway) and buffers the data in RAM as well as using the last generation "real legato link" 24-Bit DAC (which uses one of the Wadia Style digital filters). It is superior to any DAC & Transport combo I am aware of and to virtually any integrated CD Player I know (the one that betters it has a similar DAC and a more extreme mod package by Tom Evans). I had it for a long time and coming back to it.
The other is a Windows Media Center PC doing double duty as HD Jukebox, currently using a Behringer DEQ 2496 (EQ set to flat) as DAC, something more extrem is in the making (USB based, including syncronous re-clocking and avoidance of oversampling).
I know I'm not and it is work in progress.
At any extent, call me anachrophile, most of the music I like to listen to is on Vinyl anyway.
Yup, there is NO way to achieve perfect audio, or perfect anything, come to that. But in order to get certain results certain parameters are required to be met. Correct speaker directivity is one, distortion (under note of the import of the spectrum over absolute levels) and compression are others, as is reasonable pulse coherence.
But this thread is about "beaming" or directivity, so I limit myself to it. I have written more extensively about the various issues elsewhere, nontheleast here on this board, on other occasions.
"Forced directivity" is not a "Fix", it is an inherent requirement, failure to comply with this requirement requires fixes elsewhere. Please do not confuse cause and effect.
Seems quite reasonable, despite a bit wide dispersion of your speakers, sadly the "pulling off center" of the image will be very substantial in any off center location in your listening area. A centerfill channel can help no end to limit that effect, the centerfill does not the same bandwidth or level as the mains, this is not home theather.
That said, RCA style 3-Channel Stereo would be preferable for a wide sweet spot, this can be derived from 2 Channel stereo by redirecting all pure "mono" parts of the sound to a center identical to the L/R speaker while reducing the amount of the "mono" parts of the sound (but not eliminating them) in the L/R Speakers.
Auralex? I'm not sure that would be my choice, given absorbtion vs. frequency.
BTW, I suspect your 90degrees radiation pattern drawn is wildely optimistic if your speakers are normal "HiFi" style items (which they are if "Percieve" is it) something like 120degrees at least is more realistic below around 1KHz with that midrange, it has around 5" waveguide diameter, which means at best good pattern control above 1.5KHz (also visible if you look at the FR graph, below around 1.5Khz the horn (aka waveguide) unloads, leading to dropoff in SPL....
The ATC mid is rated 80 degrees spehrical dispersion, but can do so at best above around 1.5...2KHz, not below that.
Agreed. Directivity control (in other words cardiod or dipole radiators) helps no end though.
Absolutely. I find that avoiding or reducing problems from the starts has much merit, especially to trying to bolting the stable doors when the horses have already bolted.
You may or may not be aware that I have tried many methodes myself and are familiar with even more from studios, installations and with friends.
I find that all else being equal a speaker with well controlled dispersion, with good time/impulse coherence as well as low compression and low audible distortion is the solution that reduced the distance bewteen original acoustics and recording the most. In other words a speaker that fundamentally similar in behaviour to a high grade, far field monitor is a much better solution than common "HiFi" fare FOR THAT PURPOSE.
Agreed, BUT the room is ALLWAYS in the equaltion, UNLESS you live in anechonic chamber (being in one for any length of time makes me personally physically sick BTW, don't know about most others, but I know some similar cases).
Hence directivity among the key measures of quality should be counted, nay, MUST be counted, except for listening open air.
I never claimed that. However this thread is about dispersion ONLY and not about other issues (of which I am well aware).
No, but one can improve a given system to the point where the use of acoustic treatments and of EQ become optional as their invariable sonic and monetary cost (as well as te impact on decore) may become more than yielded improvement.
Again, I would suggest to reduce problems at the source instead of maximising them at the source and attempt to fix them later, sorta like in that Sound Engineers lightbulb Joke:
Q: "How many sound engineers does it take to change a lightbulb?"
A: "None, they fix it in the mix."
I for one knwo that you CANNOT fix it in the mix, you can at best excercise damage limitation, usually in a costly manner on many planes.
Sayonara
ShinOBIWAN said:
Simple. A directivity that remains constant with frequency up to around 3 - 5KHz and goes down ideally to DC and which remains either constant or preferably smoothly and slightly narrows above 3 - 5Khz, so that reflected sound follows the "house curve" (as it should, unless you want more "air" than a concert hall provides).
I suggested a DI of 6db up to the 3 - 5KHz boundary and then an increase smoothly over two octaves to a DI of 10db, in other words around 90 degree spherical -6db beamwidth below 3 - 5Khz and around 60 degrees spherical -6db beamwidth at the top of the audio band.
You may find this to be actually comparably wider at high frequencies than many common tweeters manage yet comparably narrower at lower frequencies than most tweeters will control and equally narrower than what most woofers/midranges will control through the reverbant range of the room.
ShinOBIWAN said:
They are controlled directivity for part of their usable bandwidth, in either case the waveguide is by far too small to show any effect near the lower operating frequency limit (simply look at waveguide diameter and wavelength).
They are NOT controlled diretcivity and in many ways worse than no controll at all.
ShinOBIWAN said:
(Additional) Room treatment becomes optional if the RT60 falls around 0.1 - 0.3 Seconds above 300Hz and if the Speaker has a DI of at least 6db but no more than 10db below 3 - 5KHz.
DRC is optional if the sound output from the speaker system is reasonably flat over a range of around 100Hz-10KHz (the IRT prefers 60Hz - 12.5KHz, better safe then sorry).
Both items are achievable by combining normal designs of rooms with suitably directional speakers.
ShinOBIWAN said:
With my "commercial" hat on I modify Shanling Gear, so I often have it at home. My "own" Digital players are two.
One is a heavily modded 2nd generation Pioneer DVD Player which actually plays CD and DVD Disks asyncronous (DVD Disks are always anyway) and buffers the data in RAM as well as using the last generation "real legato link" 24-Bit DAC (which uses one of the Wadia Style digital filters). It is superior to any DAC & Transport combo I am aware of and to virtually any integrated CD Player I know (the one that betters it has a similar DAC and a more extreme mod package by Tom Evans). I had it for a long time and coming back to it.
The other is a Windows Media Center PC doing double duty as HD Jukebox, currently using a Behringer DEQ 2496 (EQ set to flat) as DAC, something more extrem is in the making (USB based, including syncronous re-clocking and avoidance of oversampling).
ShinOBIWAN said:
I know I'm not and it is work in progress.
At any extent, call me anachrophile, most of the music I like to listen to is on Vinyl anyway.
ShinOBIWAN said:
Yup, there is NO way to achieve perfect audio, or perfect anything, come to that. But in order to get certain results certain parameters are required to be met. Correct speaker directivity is one, distortion (under note of the import of the spectrum over absolute levels) and compression are others, as is reasonable pulse coherence.
But this thread is about "beaming" or directivity, so I limit myself to it. I have written more extensively about the various issues elsewhere, nontheleast here on this board, on other occasions.
ShinOBIWAN said:
"Forced directivity" is not a "Fix", it is an inherent requirement, failure to comply with this requirement requires fixes elsewhere. Please do not confuse cause and effect.
ShinOBIWAN said:
Seems quite reasonable, despite a bit wide dispersion of your speakers, sadly the "pulling off center" of the image will be very substantial in any off center location in your listening area. A centerfill channel can help no end to limit that effect, the centerfill does not the same bandwidth or level as the mains, this is not home theather.
That said, RCA style 3-Channel Stereo would be preferable for a wide sweet spot, this can be derived from 2 Channel stereo by redirecting all pure "mono" parts of the sound to a center identical to the L/R speaker while reducing the amount of the "mono" parts of the sound (but not eliminating them) in the L/R Speakers.
ShinOBIWAN said:
Auralex? I'm not sure that would be my choice, given absorbtion vs. frequency.
BTW, I suspect your 90degrees radiation pattern drawn is wildely optimistic if your speakers are normal "HiFi" style items (which they are if "Percieve" is it) something like 120degrees at least is more realistic below around 1KHz with that midrange, it has around 5" waveguide diameter, which means at best good pattern control above 1.5KHz (also visible if you look at the FR graph, below around 1.5Khz the horn (aka waveguide) unloads, leading to dropoff in SPL....
An externally hosted image should be here but it was not working when we last tested it.
The ATC mid is rated 80 degrees spehrical dispersion, but can do so at best above around 1.5...2KHz, not below that.
ShinOBIWAN said:
Agreed. Directivity control (in other words cardiod or dipole radiators) helps no end though.
ShinOBIWAN said:
Absolutely. I find that avoiding or reducing problems from the starts has much merit, especially to trying to bolting the stable doors when the horses have already bolted.
You may or may not be aware that I have tried many methodes myself and are familiar with even more from studios, installations and with friends.
I find that all else being equal a speaker with well controlled dispersion, with good time/impulse coherence as well as low compression and low audible distortion is the solution that reduced the distance bewteen original acoustics and recording the most. In other words a speaker that fundamentally similar in behaviour to a high grade, far field monitor is a much better solution than common "HiFi" fare FOR THAT PURPOSE.
ShinOBIWAN said:
Agreed, BUT the room is ALLWAYS in the equaltion, UNLESS you live in anechonic chamber (being in one for any length of time makes me personally physically sick BTW, don't know about most others, but I know some similar cases).
Hence directivity among the key measures of quality should be counted, nay, MUST be counted, except for listening open air.
ShinOBIWAN said:
I never claimed that. However this thread is about dispersion ONLY and not about other issues (of which I am well aware).
ShinOBIWAN said:
No, but one can improve a given system to the point where the use of acoustic treatments and of EQ become optional as their invariable sonic and monetary cost (as well as te impact on decore) may become more than yielded improvement.
Again, I would suggest to reduce problems at the source instead of maximising them at the source and attempt to fix them later, sorta like in that Sound Engineers lightbulb Joke:
Q: "How many sound engineers does it take to change a lightbulb?"
A: "None, they fix it in the mix."
I for one knwo that you CANNOT fix it in the mix, you can at best excercise damage limitation, usually in a costly manner on many planes.
Sayonara
Hi Paul,
See this post..."How hard to clone the Orions" ..you posted in it!😉 . rather than re-type, which could take me several days..
I'm not refering to the speaker system, thats the tweeter - a Vifa XT-19. It being loaded by a waveguide.
http://www.d-s-t.com/link/vifa/data/XT19TD00-04a.htm
Now take a look at how it (remember without the waveguide) compares to the Millennium (test grp 8)
http://206.13.113.199/ncdiyaudio/ma...9mdt30start.htm
Open both drivers in two windows side by side for comparison. Look at the response between 10-20k. Notice how the XT-19 exhibits better off axis response typical of a smaller dia driver (the Millennium is very good, but more typical of a 1" dome). What this translates to often is more "air" in the treble. Now take a look at both linear and non-linear distortion. Here is where things get very, very complicated. Starting with how the testing was done, etc. then ending with -well how does this translate "sound" wise. I won't pretend to understand all the complexities involved, but the bottom line is this. The XT-19 is excellent, the Millennium is among the very, very best - if not THE best, as found by SL. It's no match at the low end, again the bigger dia. dome has the advantage here. (the dual concentric drivers like the XT are curious in that although it is a 19mm (3/4") driver, the actual diaphram is more like 1", hence (partially) the very low (unmodified) Fs of about 700Hz - but I digress). But thats where the wavguide comes in. It allows me to do many things. The 1st is the loading allows (when corrected) reduced distortion. Look at what someone intelligent enough to know what they are doing and properly record/measure the entire procedure (my polar opposite ) found:
http://www.zaphaudio.com/hornconversion.html
Then there the shaping of the directivity (response) of the driver to more closely match the directivity of the midbass (@XO) so that are more constant directivity can be achieved. There is much more, but I'm running out of gas here, so I must go. Oh yes I forgot my quick conclusion, whereas you can reduce distortion (ala WG), you cannot improve dispersion (without ill-effect, NO a phase shield IS too compromised). Whew. FYI, there is an interesting thread on the Mad board right now that you should also read. The real experts (Guru's - something I am NOT) are chiming in http://www.madisound.com/cgi-bin/di...cgi?read=349868
read the entire thread, there is good info there. I probably missed quite a bit in my description here, but it should give you some idea of my thinking WHETHER WRONG OR RIGHT. I'm out...
Bottom line, the XT-19 has superior dispersion (LESS BEAMING!) than the Millenium between say 15-20K. Is that important? To ME -yes. It can't be mimicked. You cant improve the dispersion of the Millenium between 15-20K- by say using a phase shield - without serious consequences/compromise. I refuse to get into why, its too lengthy. See thats the answer to your question Paul. There is NO ONE ANSWER. There is always going to be some sort of compromise no matter what you do.
So I chose the best possible small dia. err "dome" I could find at that time - the Vifa XT-19, because I COULD improve its low end performance/distortion/directivity,etc,etc. by using the WG.
I somewhat got my cake/eat it too. Superior dispersion @15-20k, excellent performance below. Does that make the XT-19 "better" than the Millennium? NO. It fit MY chosen compromise better. For MY loudspeaker system . That's all.
I didn't find the W22. SL did! If you or anyone can show me a better performing 8" driver within the 100-1.4k range please let me (and SL) know. Caveate - please no "I heard this and that.." - I don't care for Audiofool ramblings. I want data. Non-linear distortion data. Multi-tone tests,etc,etc. See SL's site about drive unit performance. Believe or not, he actually knows a thing or two about this stuff. There are many Audioph(oo)iles who are convinced differently. They don't need data. Just their "Golden Ears" that will betray them EVERY time there is some sort of scientific testing (ABX,etc.). Whew. I think I'm done. No time to spell check/play editor, lunch is almost over.
Cheers,
AJ
See this post..."How hard to clone the Orions" ..you posted in it!😉 . rather than re-type, which could take me several days..
I'm not refering to the speaker system, thats the tweeter - a Vifa XT-19. It being loaded by a waveguide.
http://www.d-s-t.com/link/vifa/data/XT19TD00-04a.htm
Now take a look at how it (remember without the waveguide) compares to the Millennium (test grp 8)
http://206.13.113.199/ncdiyaudio/ma...9mdt30start.htm
Open both drivers in two windows side by side for comparison. Look at the response between 10-20k. Notice how the XT-19 exhibits better off axis response typical of a smaller dia driver (the Millennium is very good, but more typical of a 1" dome). What this translates to often is more "air" in the treble. Now take a look at both linear and non-linear distortion. Here is where things get very, very complicated. Starting with how the testing was done, etc. then ending with -well how does this translate "sound" wise. I won't pretend to understand all the complexities involved, but the bottom line is this. The XT-19 is excellent, the Millennium is among the very, very best - if not THE best, as found by SL. It's no match at the low end, again the bigger dia. dome has the advantage here. (the dual concentric drivers like the XT are curious in that although it is a 19mm (3/4") driver, the actual diaphram is more like 1", hence (partially) the very low (unmodified) Fs of about 700Hz - but I digress). But thats where the wavguide comes in. It allows me to do many things. The 1st is the loading allows (when corrected) reduced distortion. Look at what someone intelligent enough to know what they are doing and properly record/measure the entire procedure (my polar opposite ) found:
http://www.zaphaudio.com/hornconversion.html
Then there the shaping of the directivity (response) of the driver to more closely match the directivity of the midbass (@XO) so that are more constant directivity can be achieved. There is much more, but I'm running out of gas here, so I must go. Oh yes I forgot my quick conclusion, whereas you can reduce distortion (ala WG), you cannot improve dispersion (without ill-effect, NO a phase shield IS too compromised). Whew. FYI, there is an interesting thread on the Mad board right now that you should also read. The real experts (Guru's - something I am NOT) are chiming in http://www.madisound.com/cgi-bin/di...cgi?read=349868
read the entire thread, there is good info there. I probably missed quite a bit in my description here, but it should give you some idea of my thinking WHETHER WRONG OR RIGHT. I'm out...
Bottom line, the XT-19 has superior dispersion (LESS BEAMING!) than the Millenium between say 15-20K. Is that important? To ME -yes. It can't be mimicked. You cant improve the dispersion of the Millenium between 15-20K- by say using a phase shield - without serious consequences/compromise. I refuse to get into why, its too lengthy. See thats the answer to your question Paul. There is NO ONE ANSWER. There is always going to be some sort of compromise no matter what you do.
So I chose the best possible small dia. err "dome" I could find at that time - the Vifa XT-19, because I COULD improve its low end performance/distortion/directivity,etc,etc. by using the WG.
I somewhat got my cake/eat it too. Superior dispersion @15-20k, excellent performance below. Does that make the XT-19 "better" than the Millennium? NO. It fit MY chosen compromise better. For MY loudspeaker system . That's all.
I didn't find the W22. SL did! If you or anyone can show me a better performing 8" driver within the 100-1.4k range please let me (and SL) know. Caveate - please no "I heard this and that.." - I don't care for Audiofool ramblings. I want data. Non-linear distortion data. Multi-tone tests,etc,etc. See SL's site about drive unit performance. Believe or not, he actually knows a thing or two about this stuff. There are many Audioph(oo)iles who are convinced differently. They don't need data. Just their "Golden Ears" that will betray them EVERY time there is some sort of scientific testing (ABX,etc.). Whew. I think I'm done. No time to spell check/play editor, lunch is almost over.
Cheers,
AJ
Kuei,
Rather than debate all this from two entirely different perspectives, I'd much rather try what you have investigated, implemented and then championed. You can talk audio but its only when you listen do you fully understand.
So to aid me in this would you be willing to suggest repectable budget parts for this experiment? Please, nothing extravagant and costly - I already have a pair of those.
Its purely to satisfy my curiosity and take comparative measurements regarding reflections within my room.
Just a couple of early questions:
I assume your using a full range dynamic driver and crossing to the horn at around 8-10khz?
Do you use a sub with that 8" dipole? If not what sort of F3 do you have in room - 40hz and what sort of SPL do you achieve with that?
Do you need dynamic EQ/limiting on the low end when playing the system loud/hard?
And what sort of roll off compensation did you use for the bass or did you increase your baffle to compensate or maybe a mixture of both? Likewise, did you use bafflestep compensation or just design the driver offset and baffle size to minimise the effects?
Hows integration between the horn and cones, since they are quite far apart or did you cross high enough to be be out of harms way?
Also any frequency/phase plots in room as well as off-axis measurements?
What are you thought's on using more dynamic drivers in a multiway configuration? 8-10" for midrange 18" for bass and horn for HF for example.
Where about in the UK are you based? Sanjian sounds like a way 🙂 but If its not far would a demo be OK by you?
Cheers,
Ant
Rather than debate all this from two entirely different perspectives, I'd much rather try what you have investigated, implemented and then championed. You can talk audio but its only when you listen do you fully understand.
So to aid me in this would you be willing to suggest repectable budget parts for this experiment? Please, nothing extravagant and costly - I already have a pair of those.
Its purely to satisfy my curiosity and take comparative measurements regarding reflections within my room.
Just a couple of early questions:
I assume your using a full range dynamic driver and crossing to the horn at around 8-10khz?
Do you use a sub with that 8" dipole? If not what sort of F3 do you have in room - 40hz and what sort of SPL do you achieve with that?
Do you need dynamic EQ/limiting on the low end when playing the system loud/hard?
And what sort of roll off compensation did you use for the bass or did you increase your baffle to compensate or maybe a mixture of both? Likewise, did you use bafflestep compensation or just design the driver offset and baffle size to minimise the effects?
Hows integration between the horn and cones, since they are quite far apart or did you cross high enough to be be out of harms way?
Also any frequency/phase plots in room as well as off-axis measurements?
What are you thought's on using more dynamic drivers in a multiway configuration? 8-10" for midrange 18" for bass and horn for HF for example.
Where about in the UK are you based? Sanjian sounds like a way 🙂 but If its not far would a demo be OK by you?
Cheers,
Ant
Konnichiwa,
I think you are better off considering "do as I say, not as I do"....
What I use at home is a strict compromise with a large nod to the beautiful young lady who for reasons inscrutible to me choose to be my wife. As such it is a workable compromise, nothing ultimate.
Budget is entierly a question of what your budget is.
The kind of stuff I use sadly does not come at a budget that is small.
Yes, that is right, maybe a little lower than 8KHz, the crossover is first order BTW. Due to the operation in "bending wave" mode at higher frequencies the driver has a much wider dispersion fairly high up than suggested by using an 8" Driver.
Here the frequency response of the base driver my special version is based upon operated with a 9V fieldcoil voltage at 0, 30 and 60 degrees:
On Axis
At 30 degrees.
Together with the dipole operation both the on and off axis response is flattened out, this driver in fact would not work well in conventional enclosures due to the varying directivity, in a plan open baffle of suitable size it works however well, as it would in a horn, where the flux boosted more to tip up the on axis response more.
I use dual 12" Pro Woofers in a small sealed enclosure, equalised, wi a 200W+ RMS Amplifier, crossover acoustically around 50Hz.
I get depending on positioning around 50Hz/-6db. The SPL with large scale classical (more difficult than modern highly compressed stuff) is high enough to **** off neigbours but not quite up to first row concert hall levels of the same piece, in the lower 90's slow average with fff. I'd like a little more, but the cost in impact on the decorum vetos this right now.
No. I also do not use any highpass.
No compensation, the baffle width is about as narrow as it can be made without incuring severe audible problems (re-flattened frequency response due to Equalisation - aka baffle step correction - nonwithstanding) for a freestanding (that is not soffit mounted) speaker, around 20".
I arranged driver offset to reasonably flatten the response.
Quite well, subjectvely speaking, some people do have to close their eyes though to avoid the visual cues to make them "hear" the tweeters.
Ever since I got this nice laptop PC I have been meaning to set it up for MLS Measurements, but other stuff has been interfering. I'll do it eventually. Measuring the system (including my subs or a pair of REL Quakes) in a number of rooms and arrangements using a pro audio 1/6th octave analyser showed a well balanced response, in room, at a number of listening positions.
You may find reviewing this post fun:
http://www.diyaudio.com/forums/showthread.php?postid=410346#post410346
If I could like I would I would not go near such toys as 8" or 10" drivers, but marital bliss compensates for compromises. maybe in the next house....
My handle and the location are actually references to the japanese anime' Sazan Eyes, highly recommended. My "evil" alter Ego KYW may live in the Himalaya in the lost land of the Sanjian, the more sociable Ego lives in London....
If you are brave enough to face London public transport, London traffic and London drivers, London ripoff prices for drinks and early closing hours and minor and other only slightly annoying hazards like terrorist bombs you are most welcome to drop by, I'm in Edmonton.
Sayonara
ShinOBIWAN said:
I think you are better off considering "do as I say, not as I do"....
What I use at home is a strict compromise with a large nod to the beautiful young lady who for reasons inscrutible to me choose to be my wife. As such it is a workable compromise, nothing ultimate.
ShinOBIWAN said:
Budget is entierly a question of what your budget is.
The kind of stuff I use sadly does not come at a budget that is small.
ShinOBIWAN said:
Yes, that is right, maybe a little lower than 8KHz, the crossover is first order BTW. Due to the operation in "bending wave" mode at higher frequencies the driver has a much wider dispersion fairly high up than suggested by using an 8" Driver.
Here the frequency response of the base driver my special version is based upon operated with a 9V fieldcoil voltage at 0, 30 and 60 degrees:
An externally hosted image should be here but it was not working when we last tested it.
On Axis
An externally hosted image should be here but it was not working when we last tested it.
At 30 degrees.
An externally hosted image should be here but it was not working when we last tested it.
Together with the dipole operation both the on and off axis response is flattened out, this driver in fact would not work well in conventional enclosures due to the varying directivity, in a plan open baffle of suitable size it works however well, as it would in a horn, where the flux boosted more to tip up the on axis response more.
ShinOBIWAN said:
I use dual 12" Pro Woofers in a small sealed enclosure, equalised, wi a 200W+ RMS Amplifier, crossover acoustically around 50Hz.
ShinOBIWAN said:
I get depending on positioning around 50Hz/-6db. The SPL with large scale classical (more difficult than modern highly compressed stuff) is high enough to **** off neigbours but not quite up to first row concert hall levels of the same piece, in the lower 90's slow average with fff. I'd like a little more, but the cost in impact on the decorum vetos this right now.
ShinOBIWAN said:
No. I also do not use any highpass.
ShinOBIWAN said:
No compensation, the baffle width is about as narrow as it can be made without incuring severe audible problems (re-flattened frequency response due to Equalisation - aka baffle step correction - nonwithstanding) for a freestanding (that is not soffit mounted) speaker, around 20".
ShinOBIWAN said:
I arranged driver offset to reasonably flatten the response.
ShinOBIWAN said:
Quite well, subjectvely speaking, some people do have to close their eyes though to avoid the visual cues to make them "hear" the tweeters.
ShinOBIWAN said:
Ever since I got this nice laptop PC I have been meaning to set it up for MLS Measurements, but other stuff has been interfering. I'll do it eventually. Measuring the system (including my subs or a pair of REL Quakes) in a number of rooms and arrangements using a pro audio 1/6th octave analyser showed a well balanced response, in room, at a number of listening positions.
ShinOBIWAN said:
You may find reviewing this post fun:
http://www.diyaudio.com/forums/showthread.php?postid=410346#post410346
If I could like I would I would not go near such toys as 8" or 10" drivers, but marital bliss compensates for compromises. maybe in the next house....
ShinOBIWAN said:
My handle and the location are actually references to the japanese anime' Sazan Eyes, highly recommended. My "evil" alter Ego KYW may live in the Himalaya in the lost land of the Sanjian, the more sociable Ego lives in London....
If you are brave enough to face London public transport, London traffic and London drivers, London ripoff prices for drinks and early closing hours and minor and other only slightly annoying hazards like terrorist bombs you are most welcome to drop by, I'm in Edmonton.
Sayonara
- Status
- Not open for further replies.