Is the box speaker a dinosaur?

[Gregadd]

IIRC somebody is advocating infinite baffles subs. I concede that is probably not a practical solution for most audiophiles.

 

[microstrip]

microstrip

Is this something you know for sure or is it a conjecture? I know for a fact that some resistors used for crossovers in passive speakers have a heat sink (Caddock comes to mind but also some Mills) ... I could be wrong but I need a fact not a conjecture to change my positions. Something I am more than willing to do.

I am far from an expert in crossover design IOW I know close to nothing .. I know however for browsing through DIY websites through that for most drivers you need more than filtering ... You also and often need to linearize either impedance of frequency response of the drivers. These increase the level of complexity and inefficiency of the crossovers, esepcially for first order which often are much larger physically that other (not a rule). In the case of woofers often a Zorbel Network is used to flatten the impedance cure of a given driver... SOme of the power that would otherwise reach the woofer is lost in the reistors of the Zorbel network. I suppose High End speakers to have the same constraints and the designers to employ similar compensating networks.. The results are to be wasted energy ... Not opinion ... facts..

Frantz,

The only way of effectively proving it to you would be computing or measuring the different power distributions in many typical passive crossovers and then debate if the chosen circuits would be representative. I could for example pick a crossover with a minimal number of resistors and get a very efficient design and another person could pick the famous Sonus Faber Extrema - a crossover in which the series element of the tweeter circuit is a very hungry power resistor. In my perspective most crossover belong to the first group and only a few exceptional are part of the second group. Most of the time we just debate in WBF the "a few" versus "most" .

I respect the DIY community and I am sometimes jealous of them - they do while we talk - but most of the time (please note the most) they design by belief than by technical reasons. In the case of resistors, most of the time people add the heatsinks just to minimize the temperature variations that could affect the resistor value due to the temperature coefficients, or just because it will looks nice. But they forget about the delays of heat propagation in solids in their argumentation about transient response. For a similar subject you should look at the net debates on the miraculous use of copper in the D'Aggostino Momemtum dissipators.

The damping and Zorbel resistors usually take very little power - they effectively ameliorate the amplifier drive by reducing the phase angle and variation of the impedance.

 

[FrantzM]

Frantz,

So what of the insertion loss in passive xovers , how is this detrimental to the sound apart from slightly lower efficiency , if efficiency is within means then it's all part of the metrics necessary for FR bandwidth control and lower distortion , it does appear you are trying to make a case for listening to your choice of IC's , so best to condemn passives .. ...

Better for you to get beyond conjecture Frantz , where's the proof pure active works , where's this digital magic you speak off , why is it better than passives ?



Please tell us where to find this fully digital active speaker , the ones i have heard are way off the pace and are not even close to their passive cousins ..

Regards ,

A. Wayne

Active crossovers allow better approximation of the mathematical functions that describe a crossover. Digital crossover are even better on this regard. TO the extent that some find them transparent read around this forum. The Digital Room Correction part.
Active crossover does not equate active speakers those are two separate concepts. A system can be made of passive speakers but with active crossovers. You said you have been living surrounded by maggies I am sure you know some people prefer to multi-amp then with active crossovers. Some highly complex speaker system use active crossovers as well .. I believe the Magico Ultimate is one of those example. There are many others.

Active speakers do work. They are not common in the High End World but if you deign to look at Andre Marc very interesting $2500 system thread, you will find several examples of Active speakers highly regarded by members here. I have a few I like personally the Dynaudio BM5A for once and the Mackie HR824 , other people with much more experience than I can possibly have in the highest expression of speaker design, one of them Gary L Koh find the Vanatoo Active speakers to be good sounding ... You OTOH have not found any you like.. Different strokes and all that .. Preferences .. Differen ears, etc .. They work and well
See above about Digital crossovers and DRC some people like (not you) for some hint of Digital magic as for Digital itis all around us whether you like ir or not is a different debate ... Different ears, strokes, preferences and all that ..

There are several exmples or active digital crossovers based systems. The Cabasse La Sphere is one of them Entirely digital Active speakers or for an actively crossed over/digital amplifiers with DRC.. Completely digital you go to Steinway Lyngdorf... tacT used to make them too. There is system from Australia or New Zealand an Apogee Like speakers with DEQX and non digital amplifiers.. The Lyngdorf site will show you example of active digital crossovers and digital amplifiers based system... The DeVialet now offers active digital crossover in the D-premier and those are the few examples to come out of my head .. Oh before I forget the Bang and Olufsen (! ) Beolab 5 .. You can pooh-pooh it all you want it is a valid example of a fully active digital speaker/system Like you told me the other day you may think about going out a little more

A little more above simple conjecture ...

@gregadd

Audiophiles solutions are not always the most practical. Multiple components in humongous Boxes, Room treatment, amplifiers that heat a room to tropical levels, cables the size of pythons are parts and parcel of what we do ... IB are just one solution they can be made to be invisible something you can't say about the better box subwoofers ... as for panel subwoofers .. I remember a friend of mine had those (Soundlabs) and they were HUGE .. Dipole subs can be panel like but also are big .. Practical? hardly ... Our quest has never been ...

Boxes speakers are here to stay .... for a while

Happy Holiday!!

 

[LL21]

So why do many of us have large box loudspeakers... It isn't about playing louder... It's about a) reproducing at a larger scale b) doing so with minimal electrical and mechanical stress that causes distortion which leads to fatigue which leads to a listening session that for lack of a better word, sucks. The very same reasons one would need and thus seek large panels.

The plain fact is that there are things box speakers do better than panels and vice versa.... From what I see now, box speakers have caught up to panels in what panels do best and may yet surpass them if the trajectory is maintained. It is panels that are making slower progress in catching up with what box speakers do well. The best panel systems still come with boxed woofers.

+1

 

[FrantzM]

Frantz,

The only way of effectively proving it to you would be computing or measuring the different power distributions in many typical passive crossovers and then debate if the chosen circuits would be representative. I could for example pick a crossover with a minimal number of resistors and get a very efficient design and another person could pick the famous Sonus Faber Extrema - a crossover in which the series element of the tweeter circuit is a very hungry power resistor. In my perspective most crossover belong to the first group and only a few exceptional are part of the second group. Most of the time we just debate in WBF the "a few" versus "most" .

I respect the DIY community and I am sometimes jealous of them - they do while we talk - but most of the time (please note the most) they design by belief than by technical reasons. In the case of resistors, most of the time people add the heatsinks just to minimize the temperature variations that could affect the resistor value due to the temperature coefficients, or just because it will looks nice. But they forget about the delays of heat propagation in solids in their argumentation about transient response. For a similar subject you should look at the net debates on the miraculous use of copper in the D'Aggostino Momemtum dissipators.

The damping and Zorbel resistors usually take very little power - they effectively ameliorate the amplifier drive by reducing the phase angle and variation of the impedance.

What's your point microstrip? I kind of got lost there ...

 

[GaryProtein]

I disagree. Materials, labor, R&D are expensive especially for relatively new companies.There is also the problem of limited product runs. Von Schwiekert says in the video I posted on the VR44 a an aluminum cabinet can cost $30-50k. . . . .

REALLY???

They can SELL an entire car for a third to half of that, which obviously requires much more manufacturing capabilities than an aluminum cabinet.

I can accept that they need to recover R&D, advertizing costs and make a profit, but don't tell me a box costs that much to build.

 

[microstrip]

What's your point microstrip? I kind of got lost there ...

My point is that you asked for proof ... And I am answering that I can not give a formal proof as you requested ...
I am putting many ... in my answer, perhaps you will not get lost this time ...

 

[FrantzM]

My point is that you asked for proof ... And I am answering that I can not give a formal proof as you requested ...
I am putting many ... in my answer, perhaps you will not get lost this time ...

Ok ... You win

 

[microstrip]

A. Wayne

Active crossovers allow better approximation of the mathematical functions that describe a crossover. Digital crossover are even better on this regard. TO the extent that some find them transparent read around this forum. The Digital Room Correction part.

(...)

There are several exmples or active digital crossovers based systems. The Cabasse La Sphere is one of them Entirely digital Active speakers or for an actively crossed over/digital amplifiers with DRC.. Completely digital you go to Steinway Lyngdorf... tacT used to make them too. There is system from Australia or New Zealand an Apogee Like speakers with DEQX and non digital amplifiers.. The Lyngdorf site will show you example of active digital crossovers and digital amplifiers based system... The DeVialet now offers active digital crossover in the D-premier and those are the few examples to come out of my head .. Oh before I forget the Bang and Olufsen (! ) Beolab 5 .. You can pooh-pooh it all you want it is a valid example of a fully active digital speaker/system Like you told me the other day you may think about going out a little more

Frantz,

There are many ways of implementing a crossover, no one has proved that the classical simple mathematical functions you are probably referring to lead to a better sounding loudspeaker than the practical functions implemented with passive components. Just because the transfer function graph looks nicer does not imply that it sounds better.

Don´t you find curious that the digital active systems that you refer to be known of sounding good cost a lot? IMHO, what separates them from lesser sounding ones is the care taken in the full implementation, not just the active digital crossover.

BTW, I have played a lot with the DCX2496 - when used above 150 Hz is does seriously affect the sound quality of my system. None of its electrical specifications can explain why.

I have nothing per se against digital crossovers. The concept is very tempting. But when I read people reporting that pure DSD sounds better than PCM at any rate I can not endorse a system that inserts a PCM 96kHz AD and DA without a lot of judging. I think that any implementation has compromises, we must debate which one allows us to have a better sounding system.

 

[Gregadd]

How far do we go to make the box work. We are addressing the problems of panels in another thread. http://www.grande-utopia-em.com/en/video.php

 

[FrantzM]

Frantz,

There are many ways of implementing a crossover, no one has proved that the classical simple mathematical functions you are probably referring to lead to a better sounding loudspeaker than the practical functions implemented with passive components. Just because the transfer function graph looks nicer does not imply that it sounds better.

Don´t you find curious that the digital active systems that you refer to be known of sounding good cost a lot? IMHO, what separates them from lesser sounding ones is the care taken in the full implementation, not just the active digital crossover.

BTW, I have played a lot with the DCX2496 - when used above 150 Hz is does seriously affect the sound quality of my system. None of its electrical specifications can explain why.

I have nothing per se against digital crossovers. The concept is very tempting. But when I read people reporting that pure DSD sounds better than PCM at any rate I can not endorse a system that inserts a PCM 96kHz AD and DA without a lot of judging. I think that any implementation has compromises, we must debate which one allows us to have a better sounding system.

Hi
Let put aside the truisms. That “ there many ways to implement a crossover “ is one of them. Modeling that is , reduce a device or phenomenon to a manageable set of equation or assumptions. That’s what is done with speakers. Any speaker from the squawker in an iPod to the Alexandria you now enjoy. The drivers also are modelized (allow me this term) . I will dispense you of the definition of a model, you know it better than most people here. We do that with speakers and guess what? These models are a pretty good indication of what these speakers can do thus the famed Thiele-Small parameters that describe very well a driver. I am certain that David Wilson and the Wilson Audio Engineers look at the Thiele-Small parameters when they have to choose drivers for their speakers, before even turning a signal on the drivers .. You will not tell me they go around listening to drivers before they read their specs (based on measurements and of course models). If it were the other way they would spend a few centuries listening to all the drivers available on this planet … The crossovers/filters are also approximation of a mathematical filter. And I can guarantee that WA make sure they have the best software available for the purpose, the software output are based on mathematical models … then they make their filters to approximate those equations and of course they listen, no problem there .. As for approximating the mathematical equation that are the filters , it is something passive crossovers simply cannot do as well as the digital ones at least in speakers. That is the power of DSP this is not an opinion it is a fact.
Now implementation is another thing s and I don’t claim that all implementation of digital filters are perfect, never did. I maintained that Digital filters approximate the mathematical functions better than analog ones and that again is a fact. Whether it sounds good to you is an opinion and I am not debating that you may simply not like anything digital some here are of this persuasion. I am not.. That is my opinion you are free to have yours.
If we are to discuss what “cost a lot” is we are in subjective territory. I however believe that the TacT that someone here use and who many think is rather transparent in that capacity is not that expensive, please allow me this subjective term in a universe where some speaker cables cost $50,000 and are trending toward $100,000. At less than $5k it is not. The Trinnov is in line with a decent non-digital, non DRC preamp at less than 7K and the Lyngdorf DP-1A under 6 K is right there in what we audiophile would call “affordable”. “Cost” a lot ? Not sure. Recently a very generous fellow by the name of Alan Jordan remarked that the extremely cheap Behringer DEQ-2496 is good sounding when everything is kept in digital ($287.85 at Amazon.com with free shipping). The DCX-2496 OTOH has to transform any input in digital and back to analog for the output. At 224.83 with free shipping from Amazon, it is likely that the Analog to Digital ( A to D) conversion has to cut several corners pun intended. Then it has to convert that into analog. No wonder it doesn’t sound that good over 150 Hz. In the bass it shines. No surprise here, that has been my observation as well, we are in agreement but let’s not forget the DEQ-2496. And you can bet your X-2 that the DIY is abuzz finding ways to make both the DCX and the DEQ objectively and subjectively better for not much more? Need I go further? No? OK
Some people say that DSD sounds better than PCM at any rate? Fine. Not my opinion. It is their right to their opinion. I can’t debate that. I am however suspect when I read however that a person find a PCM to DSD conversion better than the original. A copy better than the original!!!??l. Equal? Ok. Better? …….. !!??!?? The laws of Physics haven’t changed in spite of the End of the World that was to occur yesterday.

Happy Holiday to you ALL People!!!

 

[Gregadd]

The anti-boxes:
Available at a fraction of the price.

 

[A.wayne]

IIRC somebody is advocating infinite baffles subs. I concede that is probably not a practical solution for most audiophiles.

Agree....

Hi
Let put aside the truisms. That “ there many ways to implement a crossover “ is one of them. Modeling that is , reduce a device or phenomenon to a manageable set of equation or assumptions. That’s what is done with speakers. Any speaker from the squawker in an iPod to the Alexandria you now enjoy. The drivers also are modelized (allow me this term) . I will dispense you of the definition of a model, you know it better than most people here. We do that with speakers and guess what? These models are a pretty good indication of what these speakers can do thus the famed Thiele-Small parameters that describe very well a driver. I am certain that David Wilson and the Wilson Audio Engineers look at the Thiele-Small parameters when they have to choose drivers for their speakers, before even turning a signal on the drivers .. You will not tell me they go around listening to drivers before they read their specs (based on measurements and of course models). If it were the other way they would spend a few centuries listening to all the drivers available on this planet … The crossovers/filters are also approximation of a mathematical filter. And I can guarantee that WA make sure they have the best software available for the purpose, the software output are based on mathematical models … then they make their filters to approximate those equations and of course they listen, no problem there .. As for approximating the mathematical equation that are the filters , it is something passive crossovers simply cannot do as well as the digital ones at least in speakers. That is the power of DSP this is not an opinion it is a fact.
Now implementation is another thing s and I don’t claim that all implementation of digital filters are perfect, never did. I maintained that Digital filters approximate the mathematical functions better than analog ones and that again is a fact. Whether it sounds good to you is an opinion and I am not debating that you may simply not like anything digital some here are of this persuasion. I am not.. That is my opinion you are free to have yours.
If we are to discuss what “cost a lot” is we are in subjective territory. I however believe that the TacT that someone here use and who many think is rather transparent in that capacity is not that expensive, please allow me this subjective term in a universe where some speaker cables cost $50,000 and are trending toward $100,000. At less than $5k it is not. The Trinnov is in line with a decent non-digital, non DRC preamp at less than 7K and the Lyngdorf DP-1A under 6 K is right there in what we audiophile would call “affordable”. “Cost” a lot ? Not sure. Recently a very generous fellow by the name of Alan Jordan remarked that the extremely cheap Behringer DEQ-2496 is good sounding when everything is kept in digital ($287.85 at Amazon.com with free shipping). The DCX-2496 OTOH has to transform any input in digital and back to analog for the output. At 224.83 with free shipping from Amazon, it is likely that the Analog to Digital ( A to D) conversion has to cut several corners pun intended. Then it has to convert that into analog. No wonder it doesn’t sound that good over 150 Hz. In the bass it shines. No surprise here, that has been my observation as well, we are in agreement but let’s not forget the DEQ-2496. And you can bet your X-2 that the DIY is abuzz finding ways to make both the DCX and the DEQ objectively and subjectively better for not much more? Need I go further? No? OK
Some people say that DSD sounds better than PCM at any rate? Fine. Not my opinion. It is their right to their opinion. I can’t debate that. I am however suspect when I read however that a person find a PCM to DSD conversion better than the original. A copy better than the original!!!??l. Equal? Ok. Better? …….. !!??!?? The laws of Physics haven’t changed in spite of the End of the World that was to occur yesterday.

Happy Holiday to you ALL People!!!

Way off the pace here boss , pickup your cadence ...

Frantz,

You do realize the DSP stuff you are raving over is working on speakers with passive xovers and yes any one designing SOTA SPEAKERS ARE DOING A LOT OF LISTENING(voicing ) before TS values are even percieved ...

View attachment 7166The anti-boxes:View attachment 7165
Available at a fraction of the price.

And sound .......

 

[Gregadd]

I spoke to the designer of the Sony high end speaker and he told me that they did a lot of testing. He said an "after taste" of the cabinet was unavoidable. Here's a quote from John Atkinson's review of the smaller Sony speaker:

"He experimented with many different enclosure materials, including aluminum, but no matter how stable and solid he made a loudspeaker's cabinet, it still produced sound, ie, distortion. The important thing, therefore, would be to control that cabinet-produced sound: If the wood is going to vibrate anyway, choose a wood whose vibrations can be controlled, and are consonant with the music, not dissonant to it. "The sound coming from the cabinet should be as beautiful as possible," he told me..."

Interesting.

http://www.whatsbestforum.com/showt...t-And-should-they-be&highlight=panel+speakers

 

[FrantzM]

A. Wayne you're losing me. I stated some facts. Can you prove me wrong? Where in the world have I posted that designing SOTA speakers did not require listening? Please do show me so that I can quickly correct my statement. Straw arguments?

@greg

I am with you on the Performance to cost ratio of panels. Where we diverge is to think box speakers are dinosaurs. On the contrary I see them adapting and striving ...
Beautiful set-ups by the way especially the CLX

 

[Gregadd]

We don't really know if dinasaurs were wiped out, destroyed by a superioi species, or the victim of thier own proliferation. Buffloes roamed the plains before being hunted almost to extinction. Frantz-You and I could certainly build a speaker to spec pretty easily and with a reasonable expense. Designing a singular or cutting edge product is another matter.

Cone materials

http://www.stonessoundstudio.com.au/stone/diy_speaker_info/diy_driver_types_p3.htm

 

[Gregadd]

Horns are an option that also give us a chance to experience low wattage SET /DHTamp. I know you all are famialar with these.

 

[microstrip]

Hi
Let put aside the truisms. That “ there many ways to implement a crossover “ is one of them. Modeling that is , reduce a device or phenomenon to a manageable set of equation or assumptions. That’s what is done with speakers. Any speaker from the squawker in an iPod to the Alexandria you now enjoy. The drivers also are modelized (allow me this term) . I will dispense you of the definition of a model, you know it better than most people here. We do that with speakers and guess what? These models are a pretty good indication of what these speakers can do thus the famed Thiele-Small parameters that describe very well a driver. I am certain that David Wilson and the Wilson Audio Engineers look at the Thiele-Small parameters when they have to choose drivers for their speakers, before even turning a signal on the drivers .. You will not tell me they go around listening to drivers before they read their specs (based on measurements and of course models). If it were the other way they would spend a few centuries listening to all the drivers available on this planet … The crossovers/filters are also approximation of a mathematical filter. And I can guarantee that WA make sure they have the best software available for the purpose, the software output are based on mathematical models … then they make their filters to approximate those equations and of course they listen, no problem there .. As for approximating the mathematical equation that are the filters , it is something passive crossovers simply cannot do as well as the digital ones at least in speakers. That is the power of DSP this is not an opinion it is a fact.
Now implementation is another thing s and I don’t claim that all implementation of digital filters are perfect, never did. I maintained that Digital filters approximate the mathematical functions better than analog ones and that again is a fact. Whether it sounds good to you is an opinion and I am not debating that you may simply not like anything digital some here are of this persuasion. I am not.. That is my opinion you are free to have yours.
If we are to discuss what “cost a lot” is we are in subjective territory. I however believe that the TacT that someone here use and who many think is rather transparent in that capacity is not that expensive, please allow me this subjective term in a universe where some speaker cables cost $50,000 and are trending toward $100,000. At less than $5k it is not. The Trinnov is in line with a decent non-digital, non DRC preamp at less than 7K and the Lyngdorf DP-1A under 6 K is right there in what we audiophile would call “affordable”. “Cost” a lot ? Not sure. Recently a very generous fellow by the name of Alan Jordan remarked that the extremely cheap Behringer DEQ-2496 is good sounding when everything is kept in digital ($287.85 at Amazon.com with free shipping). The DCX-2496 OTOH has to transform any input in digital and back to analog for the output. At 224.83 with free shipping from Amazon, it is likely that the Analog to Digital ( A to D) conversion has to cut several corners pun intended. Then it has to convert that into analog. No wonder it doesn’t sound that good over 150 Hz. In the bass it shines. No surprise here, that has been my observation as well, we are in agreement but let’s not forget the DEQ-2496. And you can bet your X-2 that the DIY is abuzz finding ways to make both the DCX and the DEQ objectively and subjectively better for not much more? Need I go further? No? OK
Some people say that DSD sounds better than PCM at any rate? Fine. Not my opinion. It is their right to their opinion. I can’t debate that. I am however suspect when I read however that a person find a PCM to DSD conversion better than the original. A copy better than the original!!!??l. Equal? Ok. Better? …….. !!??!?? The laws of Physics haven’t changed in spite of the End of the World that was to occur yesterday.

Happy Holiday to you ALL People!!!

Frantz,

Although I like your repetition of a few simple known truths, my perspective is always trying to find if something can be correlated or not with high sound quality. And it is not what sounds good to me - it is what sounds good to the an appreciable majority of the audiophiles who care about high-end. IMHO this criteria is enough to establish a platform where we can have a meaningful debate, even respecting the diversity caused by one's preferences.

BTW, I am not sure that David Wilson chooses his drivers the way you refer - at less from what I could read during many years about the properties he wants his drivers to have. Assuming that "Thiele-Small parameters describe very well a driver" is nice for DIY building subwoofers, not for professional manufacturers. And yes, I have built speakers long ago this way, and even recently built a 60 litters box to measure the TS parameters of a 15" woofer - REW is of great help for these measurements.

I can not understand your comments about the DIY's and the DCX's - my apologies. And I still do not understand what the laws of Physics have to do with the copy sounding better than the original - I am assuming that being better in this question is sounding better. It seems we are now again close to the psychoacoustics you dislike so much.

 

[Gregadd]

open baffle boxless

 

[Robh3606]

Assuming that "Thiele-Small parameters describe very well a driver" is nice for DIY building subwoofers, not for professional manufacturers.

Hello Micro

LOL your kidding right. Not for the professionals!! Why do you think the professionals wasted their time on them?? For the DIY set?? When a manufacturer specs a custom woofer/midrange, aside from materials used, what parameters do you think he uses??

I'll give you hint T/S

Originally Posted by FrantzM
(...) As for the complexity of crossovers it is almost a necessity with passive speakers. This is not the subject of this discussion but it bears to remind that passive crossovers are extremely wasteful as most of an amplifier output is used to heat the resistors and inductors in passive crossovers …
Frantz,

Unhappily this myth propagates fast. Most high quality crossovers use coils of very low resistance and resistors are used in positions where energy dissipation is very low. IMHO people get this idea because most designers use large power resistors in their crossovers.

What myth is that?? You do realize that a 3db pad means that with a 100 watt peak you are dropping 50 watts across the attenuation resistors. If it's a 6dB pad it's 75 watts. Now you question why designers use large power resistors in passive networks??

Granted that there is normally no attenuation with woofers, mid ranges on the other hand, can have a good bit between band pass attenuation and any notch filters used to smooth the response.

Any attenuation or notch filters used removes amp headroom and screws up the dynamic linerarity of the system. Typically you end up with full amp power for the woofer only. The mids and highs are at best runing at -3db or quite a bit more below that. The kicker is the transient response is in the upper frequencies so if you limit the power there the system will just become strained when pushed.

As far as box speaker being dinosaurs not hardly. They are not going anywhere. Panels/dipoles will always be in the minority. Bass reflex rules, the best compromise for efficiency, power handling, bass extension and size.



Rob