A Broad Discussion of Speakers with Major Audio Luminaries

[Juhazi]

The people reading and writing in this thread should remember to differentiate between general and personal preferences. In listening tests quality/preference ratings have very wide variation between individuals, but means have trait. Outliers exist in loudspeakers and listeners and it is totally ok to like even a very coloured louspeaker, but it is not wise to declare that all others are wrong. Same with audible differencies between cables, dacs etc.

If I wanted to start loudspeaker manufacturing business, I would carefully learn and keep to principles presented by Toole and Olive.

And my humble and grateful thank you Floyd, for your lifelong work, dedication, perspective and kindness to all of us!

 

[Holmz]

The people reading and writing in this thread should remember to differentiate between general and personal preferences. In listening tests quality/preference ratings have very wide variation between individuals, but means have trait. Outliers exist in loudspeakers and listeners and it is totally ok to like even a very coloured louspeaker, but it is not wise to declare that all others are wrong. Same with audible differencies between cables, dacs etc.

If I wanted to start loudspeaker manufacturing business, I would carefully learn and keep to principles presented by Toole and Olive.

And my humble and grateful thank you Floyd, for your lifelong work, dedication, perspective and kindness to all of us!

It would be a lot easier to just take some creative writing classes, and excel at the marketing.

 

[Mnyb]

I dont don’t do ABX of speakers myself ofcourse

But I make my shortlist from well regarded speakers that seems to follow the current reasearch in speaker design.
Assuming they Kef revels ascii genelec neuman et all do have done this ( and thier other homework )

And then there are many other factors, what can I listen to locally, what’s my budget do they fit in my room ? Etc .
What can I service locally if needed or can I buy from a reputable vendor. What fits the interior design etc .

I expect reasonable good speakers from any brand to be ” good enough ” but sometimes slightly different .
But that wil even itself by room EQ and me getting used to them we seems to have an amazing ability to adapt. .

The process I adopted was to experience a cheaper model of the brand ( kef ) as a secondary speaker a couple of months , finding them good I opted for a demo on more expensive models .

I blindly bought a set of audiosmile LBM monitors for my computer thou that worked out splendidly, they are also well designed seems like speakers are being more predictable these days if you choose from brands that do follow the research ? They worked fine for the use case .

Edit: Forgot to mention sigberg , did not have the courage to order such massive speakers for a home trail as they are not in any shops . I’m sure they are also in the good speakers category

 

[krabapple]

I don't admire any of the blather here until it is at least 5 nuances deep.

 

[MKR]

We each seem to think we have cornered the market on truth. Maybe both have value, but they are very different "truths".

I can only conclude that you haven't read my book(s), AES papers or viewed my Toronto AES section YouTube lecture, because in my tests aimed at detecting resonances, the dominant flaw in loudspeakers, the actual recordings is used as test signals to energize the resonances. For me the interest in loudspeakers began when I needed technically accurate loudspeakers for my binaural hearing experiments back in 1965, a continuation of my PhD research using headphones. In my naiveté I thought loudspeakers would aim to be as accurate/neutral as electronics. I did anechoic frequency response measurements only to find that loudspeakers of that era were grossly inaccurate. Even the first blind, equal-loudness comparison tests of four loudspeakers at a time revealed a clear preference for loudspeakers that were flattish and smooth on axis. More tests revealed that resonances were what listeners complained about. The highest rated loudspeakers were the most technically accurate, which was not surprising because the same measurements described electronics, wires, etc. in the audio chain. The rest is well-documented history in AES papers and my books.

The resonances are most readily revealed using pink noise. Music was used because it was familiar and because it is the signal through which we normally would be made aware of the resonances. This provided information about how much engineering attention needed to be devoted to attenuating mechanical and acoustical resonances. Not surprisingly the spectral and temporal structure of the music determined the audibility of resonances. I would like to think it is thoroughly detailed in the publications. The results of these evaluations monotonously indicate a preference for resonance-free, timbrally neutral, technically accurate loudspeakers. Such loudspeakers transduce the electrical signal into sound with minimal change.

What do such loudspeakers sound like when playing recordings? It depends on the recordings that are mixed and manipulated in control rooms listening through (probably different) loudspeakers.
This is the origin of the circle of confusion.

Of course, so do we all. You are hearing the combined effects of variable recordings interacting with variable loudspeakers. And if you do the same tests in different rooms you will find that the small-room resonances get seriously into the act as well, to the point of even being a potential determining factor.

It sounds that what you are doing is combining the recording, the room and the loudspeaker and asking for opinions. They of course vary. If reproduced through fundamentally neutral loudspeakers it will be necessary to EQ or tone control manipulate some of them to make them appealing to fussy listeners. All assuming that small-room resonances have been attenuated.

Fortunately timbrally neutral loudspeakers are becoming common in recording studios and homes. Human judgement, preference and the effects of hearing loss among musicians and pro audio people ensure a continuing supply of "variable" recordings. Having accurate playback loudspeakers in rooms with attenuated small-room resonances is a good starting point, but "tone controls" will always be necessary I suspect.

I cannot comment on the latest cardioid loudspeakers, I am well retired, and as I have noted many times, I stopped participating in blind listening tests when I was 60 (I'm now 87) because my judgements were showing increased variability. We tracked not only the ratings of loudspeakers but the trustworthiness of listeners. My hearing thresholds then were better than expected for my age, and far better than many practicing recording engineers we have tested. And, as discussed in Chapter 16, there are many dimensions to hearing disabilities that are not revealed by threshold tests.

I would be interested in seeing spinoramas on the loudspeakers you mention - that data tells me more than I am able to hear. You obviously trust what you hear, I could for many years, but . . .

Hello @Floyd Toole … always enjoy your very thoughtful comments.

Very happy to see more discussion of cardioid loudspeakers, of which I am a big fan, especially full range cardioid (down to ~40Hz, then below this cross to monopole). My “end game” system, now being tested and fine tuned by the designer is in fact 5ch full range cardioid (l+r mains, l+r rear sides, and center all full range cardioid), with a ULF monopole sub placed just behind MLP.

I think you owe it to yourself to someday hear a full range cardioid system

Related, I wonder if you have read below paper? Would love to hear your thoughts/comments if so.

AES Convention Papers Forum » Controlling the Mode Excitation of Rooms by using Multiple Low Frequency Cardioids in Multichannel Systems

The AES Forum is where our members get to interact with each other in between conferences and section meetings.

secure.aes.org

 

[sigbergaudio]

Edit: Forgot to mention sigberg , did not have the courage to order such massive speakers for a home trail as they are not in any shops . I’m sure they are also in the good speakers category

There's a relatively new Youtuber who lives maybe 1 hour from you who has the Sarannas on loan now, if you'd like to check them out.

 

[MattHooper]

I agree, but this nuance do lack (or are at least easily missed) in the discussions sometimes. The interpretations of the measurements are also often lacking in nuance.

I agree. Often enough I see plenty of loudspeakers dismissed here as awful designs based on some level of deviation from ASR standards, and it often comes with implicit or explicit suggestions that the speaker would sound awful.

A speaker can have some deficits or liabilities yet still sound impressive in other ways.

Off the top of my head I think of the Borresen x series speakers, much excoriated around here for having some obvious frequency dips and some resonance. But both Erin and my friend (who reviewed the larger model) noticed those issues with music, but also noted some other fairly impressive qualities such as the detail, clarity, smooth highs, spacious sound staging abilities combined with excellent image focus, and “punchy bass.”

So whether either person would actually choose to live with that speaker or not (neither would), they still found some impressive attributes (and my friend said that he enjoyed plenty of music through the speaker when the defects weren’t so obvious).

I’ve mentioned before listening to the Devore O/96 which measures somewhat even on axis, but displays some resonances and some problematic off access behavior, which makes it tricky to set up in some rooms.

But with certain tracks, such as a drum solo track I used with all the different speakers I auditioned at the time, when I closed my eyes I felt closer to being in the presence of real drums with the Devore than any of the other speakers.

All sorts of loudspeakers in my room over the years, both more neutral and less neutral, have managed to leave many listeners agog at the sound “ I didn’t even know music could sound like that… it sounds so real!”

A while back I was looking at the Stereophile B&W 801D review. In B&W reviews John Atkinson has puzzled in his measurement sections more than once about the wonky nonlinear measurements. “ I’ve been to the factory they clearly have the engineering and the facilities to make more neutral loudspeakers, and yet they choose not to…why?”

In the 801 review the B&W rep was quoted:

“We don't measure or aim for a certain target in-room response. We're not tuning to a 'curve.' We develop the best engineering platform we can, and then we tune it so that we, as a group, can forget that we're listening to a pair of speakers and just get on with enjoying the listening experience. So, the target is this: Can you close your eyes and convince yourself, even for a split second, that the recording you're listening to is actually a real performer or group of performers in the room with you? If we manage to achieve that effect, we're happy.”

I haven’t heard the 801s but I’ve heard the 803D/804Ds which have the same type of signature and… I get it. I can hear the frequency sculpting and I personally would not want to own those speakers. But man do they sound clear, spacious, detailed, vivid and “boxless” from top to bottom. The sensation of “no speaker between me and the sound” and peering directly “in to the studio” to the players can be very impressive IMO. I wouldn’t want to live with them, but they are a fun place to visit (to me).

Now, of course the point often made here is that we can adjust to various colorations fairly well.

It’s really when you compare side-by-side with something “ better” that the penny drops and the deficiencies/colorations are better revealed. Best done, of course in the type of blind testing Dr Toole developed. In that case you’re likely to find out you end up preferring the more neutral lower coloration sound.

But the point of the above is simply that “less good” or “less preferred” (in blind testing) does not automatically equate to “bad sound.” A loudspeaker that does not reached the top marks in blind testing can still sound very impressive. It’s absolutely great that speaker design has progressed and now we have even better sound available . But Audiophiles have for many decades had thrilling sonic experiences even with gear that wouldn’t make the ASR-recommended list.

 

[MarkS]

A speaker can have some deficits or liabilities yet still sound impressive in other ways.

Off the top of my head I think of the Borresen x series speakers, much excoriated around here for having some obvious frequency dips and some resonance. But both Erin and my friend (who reviewed the larger model) noticed those issues with music, but also noted some other fairly impressive qualities such as the detail, clarity, smooth highs, spacious sound staging abilities combined with excellent image focus, and “punchy bass.”

You can get all that in speakers without the flaws of the Borresens for pennies on the dollar.

 

[Purité Audio]

Expensive rubbish.
Keith

 

[Floyd Toole]

I think you owe it to yourself to someday hear a full range cardioid system

Full range? The ones I know of are cardioid only at low frequencies and I cannot think of a need for it above mid frequencies. I have heard one, a prototype at Harman many years ago - it is not a new idea. None of the brands took it up. Pity.

 

[RexrothPigeon]

we tune it so that we, as a group, can forget that we're listening to a pair of speakers and just get on with enjoying the listening experience.

Implicit in this practice is an assumption that the subjective impressions of their design/tuning group are generalizable to other people. That's a completely unreasonable assumption, especially from people employed by the brand. Its unreasonableness, of course, doesn't mean it can't or doesn't sometimes work, but it sure is dumb given the empirical knowledge available to them.

 

[Floyd Toole]

But Audiophiles have for many decades had thrilling sonic experiences even with gear that wouldn’t make the ASR-recommended list.

We all have at some points in our lives. Humans adapt to what we are listening to. But, now it is not necessary to require a "layer" of perceptual accommodation unless one chooses to.

 

[MKR]

Full range? The ones I know of are cardioid only at low frequencies and I cannot think of a need for it above mid frequencies. I have heard one, a prototype at Harman many years ago - it is not a new idea. None of the brands took it up. Pity.

By full range I mean midrange down to ~40Hz. So correct, no benefit for cardioid above mid. Midrange cardioid has become more available past recent years (though still rare), but the available systems that implement LF cardioid you can count on one hand, roughly. Very rare.

Any thoughts on that article I linked?

 

[MattHooper]

You can get all that in speakers without the flaws of the Borresens for pennies on the dollar.

Of course!

I get it in my own system

 

[Jack B]

We all have at some points in our lives. Humans adapt to what we are listening to. But, now it is not necessary to require a "layer" of perceptual accommodation unless one chooses to.

Reminds me of the JBL systems from the 1970s...which undoubtedly were responsible for some of my hearing loss, but it was fun at the time.

 

[Floyd Toole]

By full range I mean midrange down to ~40Hz. So correct, no benefit for cardioid above mid. Midrange cardioid has become more available past recent years (though still rare), but the available systems that implement LF cardioid you can count on one hand, roughly. Very rare.

Any thoughts on that article I linked?

When we get into mid frequencies we are into specular reflection territory, not room modes, so it is simple directivity that matters and conventional designs do this well.

The cited paper is one I recall skimming at the time. It is fine.

It comes down to the amount of energy coupled into the room modes. Monopoles and dipoles are opposites: pressure source vs. velocity source. Cardiods are another variant. All can be made to work and it depends on how many and where they are with respect to the standing waves. Dipoles and cardioids add directional orientation as a variable, meaning that "full range" dipoles, which must be located where the "channels" are cannot fully take advantage of the cardioid feature in optimizing coupling to the modes. Cardioid subs are needed and used in the examples in the paper.

We focused on monopoles because they couple energy into high pressure regions, e.g. at the walls, where it is convenient to locate them - and, of course, they are universally available, efficient and inexpensive. The authors of this paper used one of Todd Welti's sub configurations in their example. The topic of room mode coupling is discussed in some detail in Chapter 14 and the website of the 4th edition. In all solutions matched parametric equalization can further damp modes.

There are options to attenuating room modes. The end result is what matters, and as I keep on insisting: small-room resonances are the current weak technical link in sound reproduction.

 

[Floyd Toole]

Reminds me of the JBL systems from the 1970s...which undoubtedly were responsible for some of my hearing loss, but it was fun at the time.

Here is an example: a series of loosely coupled resonances above about 1 kHz - From Figure 3.11 in the 4th edition

 

[Jack B]

Here is an example: a series of loosely coupled resonances above about 1 kHz - From Figure 3.11 in the 4th edition
View attachment 502921

Wow! As Elmer Fudd would say, "Howwible, howwible!' (Have 4th edition, just not got very far yet)

 

[Jiraya369]

In the 801 review the B&W rep was quoted:

basically more marketing speak for people who are easy to fool with sweet words. He's a rep, not an engineer, his job is to make the product appealing to the target consumers to stimulate demand and he's doing that successfully

 

[MattHooper]

basically more marketing speak for people who are easy to fool with sweet words. He's a rep, not an engineer, his job is to make the product appealing to the target consumers to stimulate demand and he's doing that successfully

Well yes of course, that’s a given.

But I wouldn’t be surprised if that is generally correct in how B&W designers are approaching things… this “ employing measurements, but ultimately designing by ear for a goal that isn’t related to strict neutrality“ approach would explain the wonky measurements.