Horn Speakers - Is it me or.......

[Newman]

Why not under 200Hz? May it be the case, that these so-called controlled listening tests showed that many people actually prefer more bass than a linear response would suggest?

I did address under 200 Hz. For some reason you omitted it when you quoted me.

Also, what is a "so-called controlled listening test"?

No speaker manufacturer does such ´controlled listening tests´, particular not blind tests asking for simple preference.

My point. But I am still not sure what is your need to equivocate any reference to controlled listening tests.

Even Harman who for a long time had had both the facility and the resources to do so, abandoned such tests a long time ago, if I have understood Dr. Olive´s remarks correctly (15+ yrs). And one of the findings he was presenting was a preference of *more* bass level than linear response would suggest. Does it mean no-one should produce neutral speakers without boosted bass anymore?

Both Toole and Olive have repeatedly stated that bass level preference is more varied among individuals than other frequency bands. Which I mentioned in my post...the part that you omitted when quoting me.

It is seemingly a common excuse for the interesting contradiction that under these ´controlled tests´, participants in the majority preferred different loudspeakers than buyers in the shops. It is nevertheless in my understanding a baseless claim.

Again with the controlled listening tests in quote marks. What's your beef?

People don´t buy expensive boutique loudspeakers which deviate from a certain ideal because of a story, look or appearance of a premium product.

Yes, they do, but they typically don't understand how those factors are unconsciously changing their perceptions of the sound quality. So, they genuinely believe it is the sound waves alone creating their impression of sound quality, not realising that the non-sonic factors can easily dominate.

Of course, if you were to ask them if they bought it for the story or appearance, they would sincerely and hotly deny it.

Otherwise it would be very easy to design ´technically righteous´ loudspeakers which beat the competition in these ´controlled listening tests´, while serving the right amount of backstory and premium feeling. According to your logic, such strategy should enable a loudspeaker manufacturer to achieve market dominance quickly.

I don't agree that my logic leads to your conclusion. They would actually have to conduct controlled listening test comparisons for your argument to hold. They never do. So how can it lead to market dominance?

But this never happened. High end loudspeakers which were designed following the findings of the ´controlled listening tests´, vastly failed when it comes to sales.

Name some, please. The vast failures.

Also, you make it sound like such speakers invariably fail, and vastly to boot. When I think of the more expensive speakers from companies like Revel, KEF, Dutch&Dutch, Genelec, Neumann, Kii Audio, Magico, I don't think of vast failures.

And guess what - people buy such non-conventional loudspeakers after extended listening tests, because they personally like the sound. You and me, we don´t have to agree to that (and with a professional background both in classical recording and loudspeaker testing, I am pretty sensitive to colorations and have a pretty good understanding which aspects of measurements correlate with which audible phenomena). But we should accept the fact that people just like what they hear and buy.

And guess what - we should also accept that their almost-inevitable auditioning process is entirely sighted listening, and hence not dominated by their preferences in relation to the sound waves themselves. As already explained if you re-read post #1113.

That said, I have listened to a lot of horn speakers which for me personally left a lot to be desired. I even noticed severe horn colorations, reverb colorations, imaging issues or restricted transparency in speakers which were from measurement point of view described as nearly perfect. While I would personally never listen to such (and almost never see them in studios, so am most certainly not alone in the recording community), I nevertheless understand what people like in them. Yes, there are certain aspects of sound quality, and personal enjoyment, which I would even say some horns do in a superior way.

I have a soft spot for good horn speakers, but it is a cognitive bias of mine, and may or may not survive a controlled listening preference test (without the strange quote marks), if I ever get the opportunity to participate in one. And from your report above, you may unconsciously have a cognitive bias against horns. (BTW what are these horn speakers with perfect spinorama and low distortion that you hated and had terrible colourations and poor imaging and transparency?)

cheers

 

[Hear Here]

That is not what the evidence is actually saying. Controlled listening tests show that people without severe hearing damage are very highly uniform in actually preferring a smooth flat frequency response between 200 Hz and 20,000 Hz. Also smooth and extended below 200 Hz, but the level varies a bit to taste.

If you went to these boutique speaker manufacturers and asked them to show you the controlled listening tests that that they have conducted that established a non-flat frequency response that is preferred by their target market… I predict with very high confidence that they will have nothing to show you. They have done no such thing. Yes, they have tuned by ear. Using sighted listening tests, which takes them straight into a spiral of imagined improvements, a spiral that never ends.

The reason they are in business is because they are fulfilling non-sonic preferences. Preferences for premium products; for unconventional products; for a certain kind of look to products; for the backstory behind a product; for the country of manufacture; for the particular kinds of things said in reviews of the product, or by other purchasers of the product talking online; for the size and weight of a product; the list goes on. And, most importantly, a preference for the cognitive biases that are created by all the above non-sonic preferences during sighted listening tests by potential buyers, and that create a perceived sonic effect that isn’t actually in the sound waves at all.

And to conclude with a paraphrase of your post’s conclusion… measurement deniers might not like it, but that is more like the real reason that they are in business, and the real market they are hitting.

Cheers

You seem to be a horn sceptic, and this probably indicates that you've never lived with one - or at least not with a good one. Whether you consider brands such as Avantgarde as "boutique" I don't know, but until about 2001, when I visited audio shows and saw crazy-looking speakers such as Avantgarde, I'd just walk past.

However, at that time I was unhappy with the speakers I'd recently bought despite having hankered after a pair for years. I was subscribing to Stereophile at the time and read this review of the Avantgarde Uno. I defy you to read this excellently-written review and not think it's a speaker you should seriously consider if the price was within your means. Take account that it was written 25 years ago and AG speakers have moved on hugely since then.

https://www.stereophile.com/content/avantgarde-acoustic-uno-series-two-loudspeaker

Please don't go straight for the Lab Results page as you'll completely miss the point of these speakers and horns in general probably. OK, I'll let you off reading the entire article if you just read the Conclusions on Page 5.

We don't take mics and PCs into concert halls and we listen to the live performance to give us pleasure. Why not consider speakers in the same light? I'm sure AG and most other brands will not put any new product on the market without extensive listening tests conducted with those with expert ears and ones who just enjoy music presented in a natural and life-like way.

After reading this article, I bought these AG Unos in 2001 and have upgraded twice since then, but have had other speakers (conventional box and electrostatics) but these exceptional horns have never failed to deliver what most of us crave - the illusion and excitement of listening in the presence of musicians.

 

[Newman]

@Hear Here , if you think your first sentence is correct, and if you think I am going to be educated by the subjective section of a Stereophile review.... I guess you have a lot to learn about me.

cheers

 

[DanielT]

The key question is what we call “affordable.” Most commercial horn speakers are absurdly expensive .......

Horns are made for DIY. To get them custom-made in size and shape plus to be able to get them at a reasonable cost.
Whether it then provides enough bass performance for the size is another question. Look at some horns and think about how many 15 or 18 inch woofers would fit in a pair of sealed or vented boxes of the same physical size, or half the physical size, and produce the same bass?

In the past with expensive amp power, low power handling and small stroke/x-max on bass drivers then horns had their justification but nowadays?

With that said. Smart DIY horn I brought up in #1027. If you make it into a shelf that is functional and blends in, you can also have a big bass horn in a smaller living room:

 

[ShadowFiend]

In the past with expensive amp power, low power handling and small stroke/x-max on bass drivers then horns had their justification but nowadays?

If you want narrow directivity (and possibly constant directivity too) from low frequency to get the absolute perception of "they are there" then horn + big woofer is the only option

 

[kharan]

Horns are made for DIY. To get them custom-made in size and shape plus to be able to get them at a reasonable cost.
Whether it then provides enough bass performance for the size is another question. Look at some horns and think about how many 15 or 18 inch woofers would fit in a pair of sealed or vented boxes of the same physical size, or half the physical size, and produce the same bass?

In the past with expensive amp power, low power handling and small stroke/x-max on bass drivers then horns had their justification but nowadays?

With that said. Smart DIY horn I brought up in #1027. If you make it into a shelf that is functional and blends in, you can also have a big bass horn in a smaller living room:
View attachment 513886View attachment 513887View attachment 513888

I've been looking at the X-Shape horn and the other stuff by the Audiohorn company, it's all quite impressive. It would have been a serious consideration if I didn't need to buy everything, learn woodworking, and find the space to build it.

 

[DanielT]

If you want narrow directivity (and possibly constant directivity too) from low frequency to get the absolute perception of "they are there" then horn + big woofer is the only option

There is something about horns. When I visit Hifi fairs, I am drawn to demo rooms with horn speakers.
It is that here and now sound that appeals to me. But it is difficult to describe what it is that appeals. Maybe it is what you mention plus possibly that highly sensitive speakers add something.
It could also be a preconceived notion about what good horns should sound like, which listening to them on site makes me imagine is the case. Or not.

It would have been interesting to switch between horns and bass boxes (sealed or ported) with several big drivers in the same listening room.

 

[Jim Taylor]

We don't take mics and PCs into concert halls and we listen to the live performance to give us pleasure. Why not consider speakers in the same light? I'm sure AG and most other brands will not put any new product on the market without extensive listening tests conducted with those with expert ears and ones who just enjoy music presented in a natural and life-like way.

Please do not confuse music PRODUCTION with music REPRODUCTION. It is true that we can listen to a live performance and enjoy it immensely. However, to gain access to that performance at a later time requires an ACCURATE recording and reproduction chain.
We do not yet have the capability of reproducing a live performance with absolute 100% accuracy ... that would take control over certain aspects which we do not (as of yet) have, including cancellation of the Second Venue.
However, we have made great strides in recording and reproduction, and we are, in my opinion, suitably close. This advancement is due to the use of measuring instruments (the mics and PCs that you mentioned) and application of scientific principles. The same goes for the advancements in medical technology, automobile engineering and communication (cell phone) technology.
In none of the latter categories I've mentioned has subjective (uncontrolled and emotional) assessment provided worthwhile input. Communication with the Mars rovers was not achieved by guess-and-by-gosh, the success of GPS wasn't achieved by opinion, nor was digital recording technology achieved by anything other than difficult, exacting scientific work.

"Extensive listening tests" are a canard, a bone thrown to the subjectivists who refuse to believe that their ears aren't as accurate or more accurate than measuring instruments. There is extensive buy-in to this misinformation in many publications and an enormous number of sites across the web. The truth is that subjective listening tests are only as accurate as human senses ... and human senses are not very accurate, being held hostage by bias. So subjective listening impressions are inherently not very accurate. Not only that, but because they are unable to be accurately transferred from one person to another, they are, at best, only useful to the unique and original listener. Someone else may find them not only not useful, but downright impedimental.

This doesn't mean that listeners do not prefer one loudspeaker design over another, including horns. Considering the almost infinite range of characteristics of the Second Venue, it is only natural that preferences exist. But that is all that they are ... personal preferences, based on emotion. Emotion is not a bad thing at all, but like the taste of an apple, the song of a bird or the beauty of a flower, it is certainly not a standard for engineering accuracy.

And neither are the opinions of reviewers in various subjective publications like Stereophile.

 

[kharan]

There is something about horns. When I visit Hifi fairs, I am drawn to demo rooms with horn speakers.
It is that here and now sound that appeals to me. But it is difficult to describe what it is that appeals. Maybe it is what you mention plus possibly that highly sensitive speakers add something.
It could also be a preconceived notion about what good horns should sound like, which listening to them on site makes me imagine is the case. Or not.

It would have been interesting to switch between horns and bass boxes (sealed or ported) with several big drivers in the same listening room.

IMHO, a 15" woofer and the capability to play bass at high levels is the biggest factor, especially considering that the consequence is that you can sit further away and therefore have a different set room acoustics while the horn still ensures that you get a decent amount of direct sound in the mids and trebles.

 

[Hear Here]

Please do not confuse music PRODUCTION with music REPRODUCTION. It is true that we can listen to a live performance and enjoy it immensely. However, to gain access to that performance at a later time requires an ACCURATE recording and reproduction chain.
We do not yet have the capability of reproducing a live performance with absolute 100% accuracy ... that would take control over certain aspects which we do not (as of yet) have, including cancellation of the Second Venue.
However, we have made great strides in recording and reproduction, and we are, in my opinion, suitably close. This advancement is due to the use of measuring instruments (the mics and PCs that you mentioned) and application of scientific principles. The same goes for the advancements in medical technology, automobile engineering and communication (cell phone) technology.
In none of the latter categories I've mentioned has subjective (uncontrolled and emotional) assessment provided worthwhile input. Communication with the Mars rovers was not achieved by guess-and-by-gosh, the success of GPS wasn't achieved by opinion, nor was digital recording technology achieved by anything other than difficult, exacting scientific work.

"Extensive listening tests" are a canard, a bone thrown to the subjectivists who refuse to believe that their ears aren't as accurate or more accurate than measuring instruments. There is extensive buy-in to this misinformation in many publications and an enormous number of sites across the web. The truth is that subjective listening tests are only as accurate as human senses ... and human senses are not very accurate, being held hostage by bias. So subjective listening impressions are inherently not very accurate. Not only that, but because they are unable to be accurately transferred from one person to another, they are, at best, only useful to the unique and original listener. Someone else may find them not only not useful, but downright impedimental.

This doesn't mean that listeners do not prefer one loudspeaker design over another, including horns. Considering the almost infinite range of characteristics of the Second Venue, it is only natural that preferences exist. But that is all that they are ... personal preferences, based on emotion. Emotion is not a bad thing at all, but like the taste of an apple, the song of a bird or the beauty of a flower, it is certainly not a standard for engineering accuracy.

And neither are the opinions of reviewers in various subjective publications like Stereophile.

The question that measurement fanatics never seem to want to answer is "Would you buy a speaker solely based on its measurements?" If the answer is No, then it follows that what it sounds like to your own ears is what really matters when parting with a large dollop of hard-earned cash. If your answer is Yes, you need help!

Where do you stand on this? Thanks

 

[Arindal]

Also, what is a "so-called controlled listening test"?

A listening tests which is claimed to be controlled, as it is a blind test with high number of participants and statistic averaging, but actually nothing is controlled that defines a useful listening test in my understanding, linking stimulus and result, judging aspects of sound quality, leading to conclusions how to optimize sound quality.

Particularly not controlled in the sense of evaluating only differences per one stimulus at the time (for example: solely frequency response modeled on the same speakers), but comparing different loudspeaker models with basically everything changing from bass to timbre. Not controlled in the sense of having no reference, like an acoustic event which had been recorded and witnessed by all listening test participants. Not controlled in the sense that participants are not familiar with recordings used (which in my understanding requires recording engineers bringing their own designated recordings). Not controlled in the sense that the main question was completely subjective, like ´Do you like loudspeaker A, B, C or D most?´ instead of rating aspects of reproduction quality, such as localization stability.

they typically don't understand how those factors are unconsciously changing their perceptions of the sound quality.

Maybe I have been doing listening tests and subjective evaluations with recording pros too often, but I was always surprised how participants in a controlled environment were sticking to ´their personal preferred sound character´, completely independent from superficial factors, with blind tests and sighted tests showing almost the same results.

Would not go so far to say that hi-fi enthusiasts have such unbiased and reliable sense of judgment, and naturally, with components offering very subtle to inaudible differences (like cables, DACs or alike), one can easily come to the conclusion that results of sighted tests are an utter delusion. Again, I don´t think this is the case with typical ´non-mainstream´ loudspeakers like audiophile horns, as differences are fairly huge, and lovers of a particular brand seem to recognize ´their sound´ easily.

Both Toole and Olive have repeatedly stated that bass level preference is more varied among individuals than other frequency bands.

Bass preference is not only more varied, it is also, according to what they were writing, further from a technical ideal of linear response. If I recall it correctly, Dr. Olive stated that if you ask participants ´do you like loudspeaker A or B more?´ the one with more and deeper bass, with statistic significance, wins.

While that is perfectly legitimate when trying to identify an average taste of the masses, and taylor products for market success (which JBL presumably managed to perfection with their portable party speakers), I see no point in taking any of these results as a target curve or reference or not even a base for judging speakers. As it is purely based on subjective opinions and solely the question ´what do you like more?´. Drawing statistic conclusions does not make the question less subjective.

They would actually have to conduct controlled listening test comparisons for your argument to hold. They never do. So how can it lead to market dominance?

You got me wrong. I was referring to a manufacturer like Harman or KEF, who have the resources to manufacture everything which is technically perfect according to their own formulated standards, and then just apply the right amount of backstory, high end pricing, fancy materials or whatever non-sonic you find necessary to convince high end customers. In my understanding, equally good stories/positioning plus research which really cracks the code of people´s subjective taste, would automatically lead to market dominance.

And seemingly they have tried in the past. It should have achieved the same market dominance they already possess in the portable speaker / party speaker / streaming speaker markets (if I think of JBL and KEF). But it did not?

Name some, please. The vast failures.

Take any expensive example of the past 25 years marketed under Harman Luxury or Synthesis brands.

When I think of the more expensive speakers from companies like Revel, KEF, Dutch&Dutch, Genelec, Neumann, Kii Audio, Magico, I don't think of vast failures.

You mix several companies which partly support the idea of just following measurements and ´controlled listening tests´ (like Revel, Neumann) and others who either publicly state that the standard measurements are vastly insufficient (like Kii, Genelec and D&D), or even one that engages in more magic audiophile stardust backstory telling than you have pictured it (Magico, who nevertheless make wonderful speakers in my perception). From how I see it, only Revel really failed, while others like KEF, Genelec or Neumann were not primarily addressing the audiophile market, but either streaming consumers or pro audio folks.

And from your report above, you may unconsciously have a cognitive bias against horns.

No, I have not, I just know the common weak spots of horns and It usually takes me 3 or 4 test tracks to expose them. I am rather sympathetic to the principle, have worked a lot with horns in my early days in pro audio, and from time to time I am positively surprised by certain products which are well-designed (I named hORNS and Strauss as examples).

what are these horn speakers with perfect spinorama and low distortion that you hated and had terrible colourations and poor imaging and transparency?

Which example do you mean, JBL M2 or K2 or Array or Everest?

 

[Jim Taylor]

The question that measurement fanatics never seem to want to answer is "Would you buy a speaker solely based on its measurements?" If the answer is No, then it follows that what it sounds like to your own ears is what really matters when parting with a large dollop of hard-earned cash. If your answer is Yes, you need help!

Where do you stand on this? Thanks

There is no need for pejorative language. Measurements serve an appropriate purpose, and the people who use them for that appropriate purpose are not "fanatics", any more than aeronautical engineers and designers are "fanatics", the people who built Hoover Dam were "fanatics", or the manufacturers of chip wafers are "fanatics".

And yes, I would choose a loudspeaker based solely on suitably extensive and accurate measurements. In fact, I no longer consider any loudspeakers for which I can not find suitable measurements. The problem for many people is not measurements, but the lack of understanding regarding those measurements. Prima facie evaluation based on slapdash analysis doesn't make the cut, yet it is in the financial interests of many manufacturers to offer only that, or even less.
One final note: it is not correct to say that I, nor anyone else, "needs help" because we do not subscribe to faulty logic and defective beliefs. That is like saying that I need help because I don't believe in voodoo, Flat Earth, phrenology or spontaneous generation.

I would appreciate future discussion based on logic and data, not name-calling. Thank you.

 

[Hear Here]

There is no need for pejorative language. Measurements serve an appropriate purpose, and the people who use them for that appropriate purpose are not "fanatics", any more than aeronautical engineers and designers are "fanatics", the people who built Hoover Dam were "fanatics", or the manufacturers of chip wafers are "fanatics".

And yes, I would choose a loudspeaker based solely on suitably extensive and accurate measurements. In fact, I no longer consider any loudspeakers for which I can not find suitable measurements. The problem for many people is not measurements, but the lack of understanding regarding those measurements. Prima facie evaluation based on slapdash analysis doesn't make the cut, yet it is in the financial interests of many manufacturers to offer only that, or even less.
One final note: it is not correct to say that I, nor anyone else, "needs help" because we do not subscribe to faulty logic and defective beliefs. That is like saying that I need help because I don't believe in voodoo, Flat Earth, phrenology or spontaneous generation.

I would appreciate future discussion based on logic and data, not name-calling. Thank you.

So, a good-measuring speaker bought solely on measurement results taken at someone else's premises will offer greater listening satisfaction than a well-reviewed speaker installed in your own home that suits your room's features and acoustics – and maybe your own ears. Wow!

Please don't take these comments too seriously - they are intended to make an important point, not as "pejorative" criticism. Let’s keep it light-hearted.

 

[Frank Dernie]

I am an old bloke who started a "HiFi" journey because I was a music lover.
All the speakers I have ever bought were not comprehensively measured simply because the technology wasn't available initially.
My first pair were DIY full range drivers (Wharfedale 8" RS/DD) in a box in 1968 bought because of availability where I lived and reputation at the time.
I used these until I bought a KEFkit 3 which was effectively the assembled baffle of their then current Concerto speaker sold for impecunious owners to put in their own box. I bought them because I liked the technology of the drivers.
I haven't changed that often, in the early days before subjective reviewers took over there was plenty of technical discussion on the engineering behind products in Wireless World, the Gramaphone and HiFi News & Record Review.

Once subjective reviewing took over the magazines I would go to listen to products that caught my fancy but rarely "got my socks blown off" and often, probably even more often than not, preferred a different product to that I went to audition!

Anyway I bought horns based on audition impression at a show then the importer and still like them but they haven't replaced, in terms of listening enjoyment, the speakers I ended up buying in the mid 1990s when I preferred them to the Wilson speakers I had gone to audition.

I would never buy based only on measurement but have used measured results to assemble an audition list.

 

[DanielT]

Speaking of horns, I heard these (horns?) at a DIY fair. Really nice bass. Bass that was felt. Big but not monster-sized. Can fit in a living room.

The DIY, FForsman, calls the bass modules a quarter-wave bandpass construction.
Or if they can be called a tapped horn/transmission line hybrid combo?
I don't know. It was good bass in any case.

More pictures and various measurements on them in this DYI thread. Starting on page 2:

HiFiForum.nu - Forsmans lek- och lärstuga - inte helt hundra. . .

www.hififorum.nu

 

[Hear Here]

I am an old bloke who started a "HiFi" journey because I was a music lover.
All the speakers I have ever bought were not comprehensively measured simply because the technology wasn't available initially.
My first pair were DIY full range drivers (Wharfedale 8" RS/DD) in a box in 1968 bought because of availability where I lived and reputation at the time.
I used these until I bought a KEFkit 3 which was effectively the assembled baffle of their then current Concerto speaker sold for impecunious owners to put in their own box. I bought them because I liked the technology of the drivers.
I haven't changed that often, in the early days before subjective reviewers took over there was plenty of technical discussion on the engineering behind products in Wireless World, the Gramaphone and HiFi News & Record Review.

Once subjective reviewing took over the magazines I would go to listen to products that caught my fancy but rarely "got my socks blown off" and often, probably even more often than not, preferred a different product to that I went to audition!

Anyway I bought horns based on audition impression at a show then the importer and still like them but they haven't replaced, in terms of listening enjoyment, the speakers I ended up buying in the mid 1990s when I preferred them to the Wilson speakers I had gone to audition.

I would never buy based only on measurement but have used measured results to assemble an audition list.

Exactly right regarding your last paragraph.

Like you, I'm an old bloke and also started with DIY projects based on Gilbert Briggs of Wharfedale's designs - I still have his books. My first was a 10" driver installed in a concrete drain pipe offering omni-directional sound - ideal in mono days. I then splashed out and tested my carpentry skills by building a pair of hexagonal Wharfedale Airedale clones - 15", 8" and 3" drivers. These were later modified with all-forward-facing KEF drivers (B139, B110 and T27) that improved the sound considerably. I then moved on to commercial products - not all successful, even the ones that measured well. Happy days!

 

[Jim Taylor]

So, a good-measuring speaker bought solely on measurement results taken at someone else's premises will offer greater listening satisfaction than a well-reviewed speaker installed in your own home that suits your room's features and acoustics – and maybe your own ears. Wow!

Please don't take these comments too seriously - they are intended to make an important point, not as "pejorative" criticism. Let’s keep it light-hearted.

An adequate suite of measurements of speakers are not measurements "taken at someone else's premises." An adequate suite of measurements are measurements taken in an anechoic chamber or, more recently, by a Klippel device. They include system sensitivity, dispersion angles at all frequencies, distortion characteristics, box resonances (if the loudspeaker is a box design) as well as other tests of SPL limits and overload characteristics. Additionally, impedance and phase behavior should be noted for the benefit of people who use amplifiers that are sensitive to such things.

It should be noted also that any loudspeaker that does not include these test results is not "well-reviewed". If you refer to a speaker being "well-reviewed" because it received a subjective review, please be advised that subjective reviews are, by and large, worthless. I say "by and large" because if the speaker (or amp) blew up and started a fire, I doubt that it would be a "well-reviewed" speaker.

The problem you pose regarding "your room's features and acoustics" deserves comment. All speakers ... those that are well-designed as well as those that are deplorable ... will be subject to the exact same acoustic effects in your room. (I think you know as well as I do that even the best of speakers can provide poor results if the agenda of the reviewer is to deliberately ruin the sound.) But a well-designed speaker (one that is judged overall as measuring better) will sound better than a poor one even under adverse circumstances ... as long as the adverse circumstances are identical.
Some people claim that there is some loudspeaker, somewhere, that has less-than-admirable characteristics, that will exhibit a response that is inverse to the characteristics to the room, cancelling them out and resulting in a wonderfully neutral and very acceptable end result.
In all my years of listening to audio gear, I have NEVER seen that brought to fruition. In all cases. the bad sound produced by a good speaker in a bad environment was better than the bad sound of a bad speaker in the same bad environment. And I had friends who believed otherwise. They spent years of their life and boatloads of money in a vain attempt to find loudspeakers with an inverse characteristic to their home acoustics.

They eventually resorted to room treatment, speakers with narrower directivity (like horns, the subject of this thread) or nearfield positioning ... and achieved greater satisfaction.

OTOH, if you're addressing ONLY the aspect of subjective preference, then your statement might apply. There is no gainsaying preference, as one of our famous members is wont to say. So if you're saying that measurements cannot predict whether you will prefer this or that speaker compared to this or that other speaker, then you are correct. For all I know, you may prefer a set of characteristics that identify a truly horrible speaker. But that destroys the validity of your comments directed to other people.

The reason for that is that one of the pillars of the Scientific Method is the concept of "reproducibility". This means that if you tell the world that you achieved a certain result, then anyone in the world, at any time now or in the future, should be able to replicate your results under the same conditions. A very simple example of this is mixing bleach and ammonia. It produces chlorine gas ... which, I might add, has an extremely offensive effect. It behaves this way today, tomorrow, next year and a hundred years from now. It does so in America, in Africa, in Asia and in Europe. Just ask Fleischmann and Pons (the cold fusion debacle) whether that is important.

Subjective preferences do not have this quality at all. Subjective preferences are personal, and apply only to the person who holds them. Other people may disagree, either a little or a lot, and it's even possible that no one on Earth will agree with your preferences. That would be statistically odd, but it might happen. And if someone does agree, it's because of happenstance and does not statistically prove anything.

The reason I point this out is that you used the phrase "listening satisfaction", which is a phrase common to subjectivist forums. I prefer to use the word "accuracy". Therefore, I would describe a speaker as most accurate, quite accurate, moderately accurate, less accurate or deplorably inaccurate.

And don't worry about my taking your comments too seriously ... I don't.

I would be delighted if you perused the information in the reference pages, and joined us in constructive discussions. Then the points you make would be of greater use to members here and the over one million visitors that view our posts yearly.

And helping them, my friend, is what is truly "Wow!"

 

[MAB]

An adequate suite of measurements of speakers are not measurements "taken at someone else's premises." An adequate suite of measurements are measurements taken in an anechoic chamber or, more recently, by a Klippel device. They include system sensitivity, dispersion angles at all frequencies, distortion characteristics, box resonances (if the loudspeaker is a box design) as well as other tests of SPL limits and overload characteristics. Additionally, impedance and phase behavior should be noted for the benefit of people who use amplifiers that are sensitive to such things.

It should be noted also that any loudspeaker that does not include these test results is not "well-reviewed". If you refer to a speaker being "well-reviewed" because it received a subjective review, please be advised that subjective reviews are, by and large, worthless. I say "by and large" because if the speaker (or amp) blew up and started a fire, I doubt that it would be a "well-reviewed" speaker.

The problem you pose regarding "your room's features and acoustics" deserves comment. All speakers ... those that are well-designed as well as those that are deplorable ... will be subject to the exact same acoustic effects in your room. (I think you know as well as I do that even the best of speakers can provide poor results if the agenda of the reviewer is to deliberately ruin the sound.) But a well-designed speaker (one that is judged overall as measuring better) will sound better than a poor one even under adverse circumstances ... as long as the adverse circumstances are identical.
Some people claim that there is some loudspeaker, somewhere, that has less-than-admirable characteristics, that will exhibit a response that is inverse to the characteristics to the room, cancelling them out and resulting in a wonderfully neutral and very acceptable end result.
In all my years of listening to audio gear, I have NEVER seen that brought to fruition. In all cases. the bad sound produced by a good speaker in a bad environment was better than the bad sound of a bad speaker in the same bad environment. And I had friends who believed otherwise. They spent years of their life and boatloads of money in a vain attempt to find loudspeakers with an inverse characteristic to their home acoustics.

They eventually resorted to room treatment, speakers with narrower directivity (like horns, the subject of this thread) or nearfield positioning ... and achieved greater satisfaction.

OTOH, if you're addressing ONLY the aspect of subjective preference, then your statement might apply. There is no gainsaying preference, as one of our famous members is wont to say. So if you're saying that measurements cannot predict whether you will prefer this or that speaker compared to this or that other speaker, then you are correct. For all I know, you may prefer a set of characteristics that identify a truly horrible speaker. But that destroys the validity of your comments directed to other people.

The reason for that is that one of the pillars of the Scientific Method is the concept of "reproducibility". This means that if you tell the world that you achieved a certain result, then anyone in the world, at any time now or in the future, should be able to replicate your results under the same conditions. A very simple example of this is mixing bleach and ammonia. It produces chlorine gas ... which, I might add, has an extremely offensive effect. It behaves this way today, tomorrow, next year and a hundred years from now. It does so in America, in Africa, in Asia and in Europe. Just ask Fleischmann and Pons (the cold fusion debacle) whether that is important.

Subjective preferences do not have this quality at all. Subjective preferences are personal, and apply only to the person who holds them. Other people may disagree, either a little or a lot, and it's even possible that no one on Earth will agree with your preferences. That would be statistically odd, but it might happen. And if someone does agree, it's because of happenstance and does not statistically prove anything.

The reason I point this out is that you used the phrase "listening satisfaction", which is a phrase common to subjectivist forums. I prefer to use the word "accuracy". Therefore, I would describe a speaker as most accurate, quite accurate, moderately accurate, less accurate or deplorably inaccurate.

And don't worry about my taking your comments too seriously ... I don't.

I would be delighted if you perused the information in the reference pages, and joined us in constructive discussions. Then the points you make would be of greater use to members here and the over one million visitors that view our posts yearly.

And helping them, my friend, is what is truly "Wow!"

Thanks for the thoughtful post. I bookmarked it.

I want to add, regular people can make their own quasi-nearfield scans, with a cheap measurement mic and a home-built turntable, and if done with care get results that correlate to a near-field scanner. And of course there are multiple threads at ASR about how to do the measurements and how the math works so that members and the ~million viewers can perhaps lift the veils of ignorance and figure out what they really like and don't like about a particular speaker, rather than useless anecdotes. And also multiple speaker reviews posted on ASR from members using the exact same home-brew SPIN method.

 

[Arindal]

I would choose a loudspeaker based solely on suitably extensive and accurate measurements.

Totally legitimate, but from vast experience with a lot of horn speakers in particularly, I have too often encountered a situation in which either audible flaws or incompatibilities with the room were not foreseeable from solely looking at measurements.

The problem for many people is not measurements, but the lack of understanding regarding those measurements.

I agree, but truly understanding measurements and predicting to a certain degree, how they would translate to sound quality or certain aspects thereof, requires decades of experience. And once you have that experience (I have btw), you become humbled by many situations in which predictions failed or actual sound quality was differing from what one might have expected solely looking at measurements.

In my understanding, it is also important to acknowledge which aspects of sound quality cannot be predicted by measurements. That said, for some aspects they are pretty useful, and I would always recommend to have experience with both worlds.

That is like saying that I need help because I don't believe in voodoo, Flat Earth, phrenology or spontaneous generation.

Judging aspects of sound quality in a controlled process, such as localization stability, bass impulse response, transparency, proximity, reverb tonality or depth-of-field, has nothing to do with esoteric beliefs. You might all call this ´subjective, but recording and mixing engineer is a respected profession, and a bunch of them can sit together, listen to their own recordings and discuss pros and cons of a speaker in question. I always found this useful to attend.

All speakers ... those that are well-designed as well as those that are deplorable ... will be subject to the exact same acoustic effects in your room. ...But a well-designed speaker (one that is judged overall as measuring better) will sound better than a poor one even under adverse circumstances ... as long as the adverse circumstances are identical.

With all due respect, I would disagree with that. There is a reason why in the recording world there are nearfield speakers, midfield speakers, main monitors and sound reinforcement systems. There is narrow dispersion and wide dispersion and spherical as well as cylindrical wave emanation, all of which can be frequency-dependent, or not, and optimized for certain room acoustics, listening distances or distance to the walls.

There are lots of well-designed speakers which only work properly under restricted conditions (such as nearfield), and fail spectacularly when used in a different room.

In speaker design and room acoustics, almost everything is a trade-off. So, yes, under certain conditions a flawed speaker from measurement perspective might be a better choice, if it delivers properties which are more suitable for the given situation in the room. The late Siegfried Linkwitz has pointed this out when opting for dipole speakers as his preferred concept to achieve constant directivity, for example.

That said, while I had many surprisingly disappointing listening sessions with (well-measuring) horn speakers, I still find them fascinating and hope that they will enable more people to overcome acoustic problems in their rooms.

 

[Hear Here]

All speakers ... those that are well-designed as well as those that are deplorable ... will be subject to the exact same acoustic effects in your room.

I don't agree with much you say, but I'll pick out this one sentence. In my own listening room, I can (and sadly have) demonstrated that your claim is totally incorrect.

A few years ago, when looking at replacing my speakers, I explored the idea of switching to electrostatics after visiting a friend's system and being highly impressed. More research led me to arrange a convincing showroom demo of the widely well-reviewed Martin Logan 13A Expression speakers - much costlier speakers than my friend's big Quads.

Stupidly, I bought these speakers without a home demo, but it was quickly obvious that these speakers could not cope with my room, whereas my ageing horns excelled there. The MLs were sold at a big loss and I upgraded to bigger, better and newer horns. I've since established the reason the electrostatics were so disappointing, despite the efforts of the UK distributor and Anthem DSP.

In my view the most important decision when considering speakers is carefully choosing the most appropriate TYPE of speaker for the room it will be working in. That's why I totally disagree with your claim above. It's demonstrably wrong, as I’ve learnt to my cost.