A Broad Discussion of Speakers with Major Audio Luminaries

[Jim Taylor]

With all due respect, it does not. I did not mean to doubt that skilled salesman and communicators can sell whatever product would not stand a controlled test. But i was referring to a not so hypothetical scenario in which a loudspeaker model superior from engineering perspective is subjectively preferred in a controlled test with great certainty, while a technically inferior model by a competitor is greatly preferred under field conditions, i.e. a sighted test, hence leading to a market dominance of the latter.

Three decades of B&W is a bright example of this happening. They made some pretty solidly engineered products in the 1990s, to a certain degree following Dr. Toole´s standards, and eventually switched to a deliberately kinked, effect-laden ´house sound´ in some generations released in the last 25 years. While taking the market by storm particularly with dealerships offering comparison tests to their customers, they were found to be greatly inferior in controlled tests undertaken by Dr. Toole´s successors at the same time, if I understood Amir's and Dr. Olive´s data correctly.

Why is that so?

To me, Dr. Toole's statement quite clearly provides an answer to your question. Perhaps I should also have bolded the words, "human nature". Your question then becomes, "Why does human nature favor non-neutral speakers under field conditions?".

I think that bias explains this quite well. In the short term, auditory bias in humans favors the obvious. For instance ... if we were walking through the woods, our brain is constantly filtering sounds to identify those which are most obvious. It is not because those sounds identify that which is most dangerous, but because those sounds identify that which is potentially most dangerous.
For that bias to be effective, it must control us rather than we control it. Rational thought cannot implement reactions quickly enough to save our lives in dangerous situations.
Nor can rational thought clearly tally the odds of a danger/no danger quality in situations of great immediacy.
Bias therefore steps in and provides us with a determination that may be wrong upon further reflection, but sufficed to favor our survival at the time.

Let's re-examine that walk in the woods. If you were to take microphones along with you, record the sounds during this walk, and play them back, you'd notice that the recordings would sound significantly different from your subjective memory. That's because your memories, as well as the initial auditory assessment, are controlled completely by bias. Instruments are not.

The first ten words of Dr. Tole's paragraph are, "Meanwhile, audiophiles trust their ears and that is a problem ...". If I were to change one word, making that statement into, "Meanwhile, audiophiles trust their biases, and that is a problem ...", would it clarify things for you?
And yes, I realize that my statement that "auditory bias favors the obvious" can be interpreted as, "Audiophiles favor the obvious." I believe that this is correct ... at least under field conditions. After all, most decisions are made under field conditions.

I see no problem here, but perhaps I am a bit thick as to the meaning of your question. Please get back to me on this.

p.s. - I had originally bolded the last sentence in Dr. Toole's paragraph. That was not because it was more important, but because I saw it as a synopsis. The whole paragraph is important, and I see it as a succinct explanation of a very important problem in audio.

 

[Floyd Toole]

a new premium car equipped with a Harman-provenance sound system which obviously had been tuned not following the science you are mentioning (really way too much bass).

Car audio is a world apart from home audio. Substantial bass boost is a requirement to compete with the background sound of a moving vehicle, and not all systems compensate for this when stationary. Some premium systems attempt to adjust it to follow vehicle speed or cabin noise levels, but not all, and not all do it well. With transducers distributed around the vehicle in locations determined by what is possible, not what is necessary, sound quality is compromised from the outset. With a thin steel and plastic "enclosure" resonances are expected, and water/moisture exposure denies the use of fluff do damp acoustical resonances. There is no stereo seat so massive signal processing is necessary to deliver a semblance of a soundstage for off-center driver and passenger. Broadband background noise masks much spatial information so multichannel upmixing and many loudspeakers are used in high-end systems to deliver a sense of spatial envelopment when at highway speeds. If a car system sounds inoffensive it is doing well, and in my experience this is possible, but not standard issue. The manufacturing of cars is such that some of the final audio system equalization and tuning has been done overnight in a parking garage in a foreign country - true story. Audio is necessary but is not a high priority. Few car platforms have been conceived with placement of audio system in mind.

Harman makes branded and unbranded audio systems for many car manufacturers, it owns the B&O car brand, and has been making B&W branded systems for years.

Harman now owns all of B&W, having purchased it as part of a bundle of well known companies at a fire-sale price. The company has been struggling for years and has been available for purchase before, including when I was with the company. We passed. When John Bowers was alive he and I communicated regularly, he sent prototypes to me at the NRCC for measurement and evaluation, he gave me a pair of his original 800 series, hosted me at a Montreaux Jazz festival, and was definitely aligned with the science. Products sounded similar. Then he died, new management took over, and there were changes. No more NRCC contact. B&W did some things well: marketing, servicing their dealers, and pushing their halo products into professional environment. Harman struggled with these things. Having numerous brands to support in both retail and automotive contexts presented challenges. These things cost money.
In the post-John-Bowers era measurements of their products indicated that there was no universal "sound" - the products were all over the place. The visually striking halo products were acoustically ill conceived - an overly large midrange combined with an unbaffled tweeter results in a large directivity mismatch at crossover, but they were a distinctive identifying feature (always an advantage), beautifully finished, and aggressively marketed, diamond tweeters and all. They weren't "bad" sounding, just not as good as they should have been at the price. A loudspeaker in this series is examined in Section 2.9.1 in the 4th edition with both technical measurements and subjective evaluations.

On the other hand, if controlled tests are mainly about subjective preference, I would expect the products succeeding in such environment, also to be preferred under sighted conditions typically found in living rooms and at dealerships, at least in the majority of cases. Even if conditions are non-ideal and listeners are biased this or that way, why wouldn't they choose the very same speaker as best-sounding they had preferred under controlled conditions? It is kind of counterintuitive to me assuming the majority would buy speakers they would not prefer.

Sorry but this is demonstrably wrong. An example: Toole, F. E. and Olive, S.E. (1994). “Hearing is believing vs. believing is hearing: blind vs. sighted listening tests and other interesting things”. 97th Convention, Audio Eng. Soc., Preprint 3894. There both experienced and inexperienced listeners of both sexes demonstrated that they heard less when listening sighted, ignoring not subtle, but easily heard differences. It is described in the books as well (Chapter 10 in the 4th edition). Blind evaluations are done in many contexts for good reason.

Listening in a sighted situation, in stereo, in what is likely to be an unfamiliar acoustical setting, even familiar music won't save the day - a biased result is inevitable. But it is human nature to believe what they think they hear at any time and place is the real "truth". See slide shows 1 through 3 on the book website.

And although I fully support the idea of science and competent engineering, I have my doubts that even a noticeable, increasing minority of the market is really following.

As I said in my last post to you, it doesn't matter whether anyone notices or not. The masses are not going to seek out measurements when selecting loudspeakers, but a few, like participants in this forum probably will have a look. Those who need persuasion are manufacturers, and there again, not all will abandon the old ways. All that can be done is to present the facts, and that has now become much easier.

Today shoppers are increasingly likely to end up with relatively neutral loudspeakers, and that is a good thing. Competent engineers in responsible manufacturers are seeing to it, but not all manufacturers are responsible so-o-o-o. It is still early days in the utilization of the ANSI/CTA standard - the spinorama - but it pleasantly surprises me the extent to which it has been accepted. Any shift in that direction is good.

Some people believe they hear differences between well designed amplifiers, DACs and wires. Science, measurements, A/B/X subjective evaluations show otherwise, but consumers will get good sound no matter what they think they "prefer". Competent engineers in responsible manufacturers have seen to it for decades.

Scientific findings have never been the go-to guidance for the masses, so being in a minority is the norm. 'Twas ever so.

 

[Arindal]

Substantial bass boost is a requirement to compete with the background sound of a moving vehicle, and not all systems compensate for this when stationary.

I am aware of that, but I can say that I am referring to modern premium BEV which need almost no bass boost because they are so quiet even when being driven really fast. And the bass boost of this particular system is obviously way way more emphasized than any kind of compensation even for an outdated ICE car.

Harman makes branded and unbranded audio systems for many car manufacturers, it owns the B&O car brand, and has been making B&W branded systems for years.

That is why I have written Harman-provenance, not branded Harman/Kardon. Properly (heavily) equalized, the systems I have heard are excellent. So my conclusion would be some kind of extreme ´house sound curve´ is at play here, in the region of +10 or +12dB bass boost as well as some brillance/treble boost.

he gave me a pair of his original 800 series, hosted me at a Montreaux Jazz festival, and was definitely aligned with the science. Products sounded similar. Then he died, new management took over, and there were changes.

Interestingly, the really linear B&W speakers with which I am familiar with, both in pro audio and consumer audio, were dating back to the 1990s, so after John Bowers had passed. Back in the days they were pretty popular among many recording engineers and mastering studios I have worked with.

The idea that something has changed and some ´house sound curve´ deviating from the technical ideal was implemented, as well as questionable marketing claims arising, only caught my attention somewhen between 2000 and 2010. Was not aware of them struggling in this period, rather noticed the opposite, i.e. B&W being sold several times for astonishing sums, last time in 2022.

The visually striking halo products were acoustically ill conceived - an overly large midrange combined with an unbaffled tweeter results in a large directivity mismatch at crossover,

Fully agree to your concerns regarding directivity misconception, but interestingly the first and second Matrix 801 series, which were pretty popular in studios back in the days, offered a similar baffle-less design and the directivity was broad and a bit inconsistent, yet not completely off (Stereophile measurement):

Personally never really liked them, but some very respected recording engineers, among them Grammy winners and others, have explained to me that they prefer the advantage of no edge diffraction problems, very consistent reflection tonality and no overall dullness in the reverb field under studio conditions to what was available at the time.

A loudspeaker in this series is examined in Section 2.9.1 in the 4th edition with both technical measurements and subjective evaluations.

If I recall it correctly, that was already one of the models with such ´house sound curve´, so we agree on this point.

B&W did some things well: marketing, servicing their dealers, and pushing their halo products into professional environment. Harman struggled with these things. Having numerous brands to support in both retail and automotive contexts presented challenges. These things cost money.

I did not come to the conclusion that lack of money, brand reputation, pro audio presence or dealerships were really the key problem of Harman brands in high end audio. Rather the opposite, as JBL and Infinity were among the best known audio brands of all in the heydays of hi-fi. And pretty successful in their market segments.

JBL is still, maybe even the No. 1 brand among younger folks. Ironically, they are seemingly successful with products showing a pretty pronounced ´house curve´ instead of linear frequency response. At least you come this impression whenever you look at people in the park or on the beach listening to various portable speakers from Charge 6 to huge Partybox models. As bass-heavy as the in-car systems, far from any theoretical ideal of linear response, I would say.

Listening in a sighted situation, in stereo, in what is likely to be an unfamiliar acoustical setting, even familiar music won't save the day - a biased result is inevitable.

Again, no disagreement from my side. It just does not explain why the preference in a loudspeaker A/B comparison would completely and consistently reverse from the technically superior model to the inferior one, just because the controlled blind test was exchanged by a sighted one with all parameters (like room, placement and music) staying the same.

I have been taking part in such tests a long time ago, and conducting some myself. I found the blind vs. sighted results to be surprisingly consistent. Admittingly, this took place with mainly recording engineers and studio folks as listeners, using mainly their own recordings. So one might argue they recognize their preferred sound ideal even under blind conditions (which was in most cases the linear variant anyways, so congruent with your findings, but this stayed completely consistent in sighted tests).

The masses are not going to seek out measurements when selecting loudspeakers, but a few, like participants in this forum probably will have a look. Those who need persuasion are manufacturers, and there again, not all will abandon the old ways. All that can be done is to present the facts, and that has now become much easier.

Fully agree and support that idea.

My only point is the way this idea is communicated today and to whom. Would assume that there is a vast crowd of potential buyers who are pretty confused but interested in better sound quality and would be willing to invest into better sounding products, but the old ways of communicating the importance of measurements and specs, as tried in the 1970s and 1980s, are just not attractive to them.

 

[MarkS]

Dr. Toole, a poster here has claimed that you would agree with the following statement:

"Sighted listening is wholly and always unreliable.”

Do you agree with that statement?

And if you do, then can you please explain to me why I should prefer well-measuring speakers over poorly-measuring speakers. I listen to my speakers sighted. If I cannot reliably tell the difference between their sound under sighted conditions, why should I prefer well-measuring speakers?

Thank you.

 

[Arindal]

why I should prefer well-measuring speakers over poorly-measuring speakers. I listen to my speakers sighted. If I cannot reliably tell the difference between their sound under sighted conditions, why should I prefer well-measuring speakers?

If you cannot tell the difference in either case, maybe you are just lucky and can live with cheaper, rather poorly-measuring speakers anyways?

I would neither subscribe to the statement on sighted listening you have suggested, nor have I understood Dr. Toole´s explanation that way. But I can clearly tell you that better, well-measuring speakers, exhibiting less audible flaws in whatever controlled tests, are in my opinion a better choice if you are going to listen to a variety of differently sounding recordings in the future.

 

[NTK]

Dr. Toole, a poster here has claimed that you would agree with the following statement:

"Sighted listening is wholly and always unreliable.”

Do you agree with that statement?

And if you do, then can you please explain to me why I should prefer well-measuring speakers over poorly-measuring speakers. I listen to my speakers sighted. If I cannot reliably tell the difference between their sound under sighted conditions, why should I prefer well-measuring speakers?

Thank you.

The key concept is "consistency".

 

[Floyd Toole]

I am referring to modern premium BEV

You are correct, the amount of boost should relate to the background noise in any specific vehicle. If not done, the engineers need disciplining. Common sense.

some very respected recording engineers, among them Grammy winners and others, have explained to me that they prefer the advantage of no edge diffraction problems, very consistent reflection tonality and no overall dullness in the reverb field under studio conditions to what was available at the time.

I have spend many hours discussing this kind of thing with recording engineers and routinely am puzzled by some of the arguments. It is sometimes as though the music and the musicians had little to do with a record's success in the charts; it was the monitor speaker, or the console, or some other technical thing. I absolutely agree that recording and mastering engineers share in the credit, and that the tools they use are part of the process, and that they had a choice in the tools.

I won't repeat the details here because they are in all my books and AES papers from 1985/86, but when Canadian Broadcasting Corporation recording engineers participated in double-blind, equal loudness, multiple comparison evaluations of loudspeakers to choose new monitors their opinions agreed with those of audiophiles, and both groups preferred timbrally neutral, resonance- free loudspeakers. The best ones in the test at that time were consumer products, and the professionals were surprised by their choices, some demanding retests using their own master tapes - no difference. A couple commented that they had never heard such good sound before. Sadly, several of them exhibited quite high variations in their judgments when hearing the same sounds repeated, and this correlated with hearing loss, an occupational hazard in pro audio. The lowest rated product in the test was a UREI 811B monitor.

With respect to the B&W I wonder how your engineers were able to hear the absence of "edge diffraction problems"? Obviously they saw curved surfaces and assumed it. I'll bet nobody actually did anechoic measurements to confirm it. In fact the unbaffled tweeter is a problem, not a virtue. This is why most neutral loudspeakers use baffles and horns/waveguides on tweeters to match directivites at crossover. Hearing "constant reflection tonality" is possible only if the loudspeaker exhibits relatively constant frequency dependent directivity, which this one patently does not. They needed measurements in order to understand what they were listening to.

Here is a quote from the 4th edition: "An interesting fact is that the B&W 802N is found in some high-profile music recording facilities and scoring stages. Some recording engineers believe it to be especially useful for classical music. It has been said that the sagging output around 3 kHz is compensation for excessive brightness in strings − the result of placing microphones above the violins where they project energetic high frequencies that are not heard in that proportion by live audiences (Meyer, 2009). If this is the case, the remedy is to equalize the signal, not to use monitor loudspeakers with a compensating error, thereby contributing to the circle of confusion. In any event this is a non-neutral loudspeaker being used in situations where neutrality should be a requirement."

Award winning mixes have also been attributed to the use of NS-10s and Auratone 5Cs, which are not on anyone's list of loudspeakers to listen to for pleasure. Here is a quote from the Auratone website: "In 1982 Michael Jackson’s Thriller album, the best selling album of all time, was mixed on the 5C’s with Quincy Jones and Bruce Swedien. Bruce said that “80 percent of the mix was done on the 5C’s” and that Quincy calls them the “truth speakers."

The popular UREI 800s were remarkably like the Auratones, but with bass, and lots of rock was mixed on them. They were so bad off axis that the "dead end" was invented to absorb the offensive sounds. I attach measurements - Figure 3.13 in the 4th edition.

I discuss the "transfer" of recorded arts from creators to consumers in the 4th edition, with a parade of measurements from all categories of loudspeakers in slide show 6. It is evident that mixers are trending to neutral loudspeakers, in part because that is what major manufacturers are offering them. If any want to hear what their mix sounds like through an NS-10M, a B&W or anything else, which is not unreasonable, it can be accomplished by appropriately equalizing a neutral loudspeaker. The fact that the current trend is to close/near-field listening or listening in reflection reduced acoustically dead rooms makes the task easier - the direct sound dominates. The remaining problem is that consumers don't listen in the near field, so that cannot be the final evaluation.

 

[Floyd Toole]

Dr. Toole, a poster here has claimed that you would agree with the following statement:

"Sighted listening is wholly and always unreliable.”

Do you agree with that statement?

And if you do, then can you please explain to me why I should prefer well-measuring speakers over poorly-measuring speakers. I listen to my speakers sighted. If I cannot reliably tell the difference between their sound under sighted conditions, why should I prefer well-measuring speakers?

Thank you.

When discussing subjective evaluations - opinions - one never speaks in absolutes. So I would like to think I never said that. The truth is that sighted listening opens the door to countless non-auditory influences on the judgement forming process. This is well discussed early in the 4th edition.

The most trustworthy evaluations are those in which there is only one variable: the loudspeaker. The double-blind, equal-loudness, multiple (3 or 4) loudspeaker comparison method has been shown to be the most revealing and the most reliable.

The widely used "take it home and listen to it" method is the least reliable.

You said: "If I cannot reliably tell the difference between their sound under sighted conditions, why should I prefer well-measuring speakers?" I cannot respond to this statement because we need to know the measurements on the loudspeakers you were comparing. If there is a large difference and you cannot hear it, perhaps you have a hearing problem. Differences can be present, sighted or blind, whether one responds to the acoustical evidence is partly dictated by non-auditory factors.

 

[MarkS]

Oh, I easily hear the difference. I recently bought AsciLab C6B speakers (which measure superbly). My previous speakers were GoldenEar Triton Sevens (which measure decently but not great). Since I'd had the GoldenEars for several years and had presumably adapted to their sound, I was expecting the AsciLabs to sound different but not better, at least at first. I was wrong. As soon as I set up the AsciLabs (and ran RoomPerfect room correction, as I had for the Tritons), they immediately sounded better (on a recording of a string quartet that I have heard play live), compared to my memory of the GoldenEars. I did not do any back-and-forth switching, and of course this was all done sighted.

There are posters here who claim that I am deluding myself, that it is flatly impossible to trust such a sighted listening prefenence, that sighted listening is never a valid way to judge speakers.

I disagree. I think the AsciLabs are clearly better than the GoldenEars, and that I could hear this within a few minutes of sighted listening.

I'd be very interested in your thoughts on this, thank you.

 

[Floyd Toole]

Oh, I easily hear the difference. I recently bought AsciLab C6B speakers (which measure superbly). My previous speakers were GoldenEar Triton Sevens (which measure decently but not great). Since I'd had the GoldenEars for several years and had presumably adapted to their sound, I was expecting the AsciLabs to sound different but not better, at least at first. I was wrong. As soon as I set up the AsciLabs (and ran RoomPerfect room correction, as I had for the Tritons), they immediately sounded better (on a recording of a string quartet that I have heard play live), compared to my memory of the GoldenEars. I did not do any back-and-forth switching, and of course this was all done sighted.

There are posters here who claim that I am deluding myself, that it is flatly impossible to trust such a sighted listening prefenence, that sighted listening is never a valid way to judge speakers.

I disagree. I think the AsciLabs are clearly better than the GoldenEars, and that I could hear this within a few minutes of sighted listening.

I'd be very interested in your thoughts on this, thank you.

As I said if the only variable is the loudspeaker the chances of valid judgements increases. If only the sound is being judged, the validity is even better. If non-acoustical factors are involved the chance of an unbiased judgement is reduced. You say that in this particular test you could be unbiased - maybe so. But in this case, perhaps the difference in sound was great enough to overcome any residual bias. The AsciiLabs is an excellent product, showing signs of very competent engineering - which no doubt is why you bought them - bias no.1, but a good bias

 

[sigbergaudio]

Oh, I easily hear the difference. I recently bought AsciLab C6B speakers (which measure superbly). My previous speakers were GoldenEar Triton Sevens (which measure decently but not great). Since I'd had the GoldenEars for several years and had presumably adapted to their sound, I was expecting the AsciLabs to sound different but not better, at least at first. I was wrong. As soon as I set up the AsciLabs (and ran RoomPerfect room correction, as I had for the Tritons), they immediately sounded better (on a recording of a string quartet that I have heard play live), compared to my memory of the GoldenEars. I did not do any back-and-forth switching, and of course this was all done sighted.

There are posters here who claim that I am deluding myself, that it is flatly impossible to trust such a sighted listening prefenence, that sighted listening is never a valid way to judge speakers.

I disagree. I think the AsciLabs are clearly better than the GoldenEars, and that I could hear this within a few minutes of sighted listening.

I'd be very interested in your thoughts on this, thank you.

I guess the point isn't that it's impossible to hear the difference between a good and bad speaker in a sighted listening situation. But it introduces uncertainty into which perceived differences are attributed to bias, and which perceived differences are attributed to the actual sound.

 

[RexrothPigeon]

If you cannot tell the difference in either case, maybe you are just lucky and can live with cheaper, rather poorly-measuring speakers anyways?

One thing ASR shows on a regular basis is that price and accuracy aren't reliably linked. Those "rather poorly-measuring speakers" are just as likely to be more expensive as they are to be cheaper.

 

[MattHooper]

On sighted listening / auditioning of loudspeakers…

I did a very large speaker search a few years back in which I listened to a very wide variety of “high end” loudspeakers. This included ASR -approved speakers like Kii Audio actives and Revel, to ASR-disparaged brands like Devore, and plenty of speakers in between.

What I perceived in many instances (my impressions) were many speakers with different varieties of colorations. My perception could’ve been of course inaccurate for all the reasons sighted listening can be inaccurate.

However, there were times when I heard a loudspeaker first, perceived certain colorations, and then later was able to find measurements (for instance in stereophile, soundstage etc). And some relevant characteristics I perceived in auditioning the speakers showed up in the measurements. Off the top I’m thinking of the Paradigm Persona speakers which I perceived as generally even but with a sharp edge in the highs that grew fatiguing during the audition to the point I had to turn down the volume over time. I left earlier than I otherwise would have and wrote them off my list. And that’s pretty much how they measured in stereophile - on axis generally even but with a serious rise in the highs starting around 6k to a 5DB peak at 10 K. Enough to bother my ears anyway.

Likewise I spent a while listening to plenty of my tracks on the PMC FACT 8 speakers and perceived them as having a bit of an emphasized top end giving an impression of “ shine” to cymbals and a slight accentuation of upper frequency detail, but I was most put off by a lack of warmth somewhere in the lower mid or upper bass, which made them sound a bit, mechanical, and tending to emphasize the artificial nature of recordings.

Later, I saw the stereophile review, Kal Rubinson also noted this lack of warmth in the same region before he saw any measurements, and the measurements track quite well with what I perceived - bit of a dip around the upper bass and a rising high-end peaking at 10 K.

Or there’s the Rhaido speakers which in a store audition disappointed me by sounding obviously coloured - a bulging upper bass - all sorts of my favourite tracks felt obnoxious in the bass region and lacking presence and dynamics in the upper mids. I found out later from measurements this was their “house sound.”

Or these current B&W 804 D4 speakers that sounded to me under sighted conditions “ like they measure” in terms of that roller coaster through the mids and high frequencies.

Along those lines, I have a friend who writes reviews for soundstage and so I’ve been joining him to listen to the loudspeakers under review for decades, both of us forming our impressions of the sound before the speakers are sent off to be measured at the NRC.

Many times characteristics perceived in the informal listening seemed to show up in the measurements - lean bass here, obvious suck out or peak in the frequency response there etc.

What about speaker set measure well?

I auditioned the Revel Performa F228Be more than once, at two different shops. What I perceived was easily among the best of the many loudspeakers I auditioned.

The main trait that stuck out to me was a sense of general neutrality and an even balance from top to bottom, with great imaging and sound staging, nothing stuck in the speaker, and they sounded smooth off axis as well. It was damn hard to pick any nits in the sound.

That seems to track quite well with how they actually measure and perform in blind listening tests.

But… I was aware that Revels measure well. So that could’ve been some form of bias in my perception. But on the other hand… it seems plausible to me that since that’s how they actually sound, I was picking up on the actual General characteristics of the loudspeaker. (it also happens that the first time I encountered the Revels I was surprised that they sounded off, a bit lumpy and also somewhat coarse highs. But that pair was set up in another store suboptimally - one of the speakers near a huge wall length glass panel, the other too close to the back wall. So my knowledge Revels measure well didn’t seem to overcome a poor set up in that instance).

Or the several times I auditioned the Kii Audio Three active stand mount, and BXT system. Again my overriding impression was of basic neutrality, excellent control from top to bottom in the frequencies, and a really impressive “disappearing speaker” act in terms of nothing sounding like he was coming directly from speakers. The bass was particularly impressive - the speakers even with the BXT system, were close to the wall behind them where I would expect most speakers to sound boosted in the bass, but there was no such problem and the bass was amazingly clean and even and tight.

That is of course exactly the type of benefits ascribed to the Kii’s cardiod radiation design.

Was my perception of these characteristics - which align well with benefits of the design as well as how they generally measure - just coincidence or a bias effect?

Can’t say for sure it wasn’t.

But it also seems to me plausible that I was perceiving some of the real sonic characteristics of that speaker, and that those sighted listening impressions had some level of accuracy.

And if one isn’t going to actually perceive the benefits of that active cardioid design in sighted listening, what would be the point of buying it for those performance benefits in the first place?

I have no reason to believe that I would perform any “better” in the type of blind speaker listening tests we talk about here. There is no doubt that I would’ve experienced various levels of bias effects during my long time as an audiophile owning and auditioning speakers. So I’m just trying to still make sense of the type of experiences above, where informal listening impressions seem consistent with measurements, keeping that context in mind.

I’ve noted before also that Erin of Erin’s Audio Corner derives his listening impressions under sighted conditions before he measures them because he doesn’t want to be influenced by the measurements (while also admitting that blind testing would be even better). Yet it seems much of the time Erin’s sighted listening impressions catch all sorts of characteristics and colorations (or lack of colorations) that are corroborated afterwards with his Klippel scanner measurements.
(unless we want to disbelieve Erin’s claims about his reviewing method… which of course anyone is free to do).

Of course, none of the references to sighted impressions above are suitable for scientific certainty.

But as a practical every day matter, I think some impressions under sighted conditions can be useful - understood with lower confidence levels and appropriate caveats.

Otherwise, again, the relevance of blind listening and measurements to real-world listening conditions becomes suspect.

 

[MarkS]

I guess the point isn't that it's impossible to hear the difference between a good and bad speaker in a sighted listening situation. But it introduces uncertainty into which perceived differences are attributed to bias, and which perceived differences are attributed to the actual sound.

The data presented in Dr. Toole's book clearly shows that sighted and blind listening results are correlated. For mysterious reasons, some here continue to believe (against all evidence and against all expert statements) that this is not the case, and that sighted listening is completely worthless.

If it was true that sighted and blind listening results were completely uncorrelated (this is, again, not true), then blind listening would be useless for evaluating speakers that one intends to listen to sighted. It would mean that a speaker's visual appearance completely overwhelms its sonic qualities, and so you should choose speakers based entirely on which one looks nicest to you.

I completely accept that sighted listening is less reliable than blind listening. But it is not completely unreliable. If it was, no one could tell the difference between a fully functioning Revel Salon 2 and a dummy version (visually identical) that was in front of a Sonos Era 100 that was actually producing the sound.

 

[RexrothPigeon]

I completely accept that sighted listening is less reliable than blind listening. But it is not completely unreliable.

The thing about sighted listening is that while technically you are right, there is no way of knowing when it reliable or even whether it is.

This is why there's so much resistance to sighted listening reports here. It's so seductive for people to sorta decide that their own sighted listening is the exception. But there's no way for them or us to know, so it's just a game of trust me bro. I find that kind of thing tedious and uninteresting, but obviously not everyone agrees since most hifi discourse relies on it exclusively.

 

[Jim Taylor]

The thing about sighted listening is that while technically you are right, there is no way of knowing when it reliable or even whether it is.

This is why there's so much resistance to sighted listening reports here. It's so seductive for people to sorta decide that their own sighted listening is the exception. But there's no way for them or us to know, so it's just a game of trust me bro. I find that kind of thing tedious and uninteresting, but obviously not everyone agrees since most hifi discourse relies on it exclusively.

Many posters in this thread have mentioned that either 1) they like one speaker more than another, or 2) they can definitely hear that one is better than the other.

These seem to be statements of subjective preference, not at all comments on accuracy, and not at all concerned with scientific rigor. If people would simply say that they can differentiate one speaker from the other, or that they definitely like one speaker compared to the other, then I would accept their statements as completely plausible.

Unfortunately, they do not express themselves in such a manner.

 

[RexrothPigeon]

2) they can definitely hear that one is better than the other.

I've taken the liberty of bolding the part that is often in conflict with accepted science. The part you bolded is, as you said, is completely unobjectionable

 

[sigbergaudio]

The data presented in Dr. Toole's book clearly shows that sighted and blind listening results are correlated. For mysterious reasons, some here continue to believe (against all evidence and against all expert statements) that this is not the case, and that sighted listening is completely worthless.

If it was true that sighted and blind listening results were completely uncorrelated (this is, again, not true), then blind listening would be useless for evaluating speakers that one intends to listen to sighted. It would mean that a speaker's visual appearance completely overwhelms its sonic qualities, and so you should choose speakers based entirely on which one looks nicest to you.

I completely accept that sighted listening is less reliable than blind listening. But it is not completely unreliable. If it was, no one could tell the difference between a fully functioning Revel Salon 2 and a dummy version (visually identical) that was in front of a Sonos Era 100 that was actually producing the sound.

I am not trying to say that it is uncorrelated, I have similar experiences as @MattHooper and Erin that what I experience with a speaker often seem to correlate well with what is actually happening. I do a lot of sighted listening of my own designs during development as well, walking back and forth between listening and measurements. When something feels off during listening, I try to find the source in the measurements, and most of the time if some frequency range feels off, lo and behold that's supported in the measurements. But that's of course a more analytical approach and situation then going to a store to listen to a pair of speakers after reading a review.

I've also experienced the opposite, that I do a change in the tuning, and afterwards I hear that yes, that really worked well. Then later I find out that I never loaded the change to the speaker. So.. Our ears certainly work, but they're overriden by the brain sometimes.

 

[Jiraya369]

I've also experienced the opposite, that I do a change in the tuning, and afterwards I hear that yes, that really worked well. Then later I find out that I never loaded the change to the speaker. So.. Our ears certainly work, but they're overriden by the brain sometimes.

i had something like this happen whilst mixing, i was mixing and used an eq for bass and the muddiness went away

2 mins later i went and saw i had turned the mixing track off so everything was as is. Placebo is scary! At least I know i can get affected by it and dont think i have golden ears lol

 

[MaxwellsEq]

For mysterious reasons, some here continue to believe (against all evidence and against all expert statements) that this is not the case, and that sighted listening is completely worthless.

You are misunderstanding or miss-characterising what people have stated.

If you take part in a blinded listening test using complex pop music via several speakers on which you have no financial or purchasing interest, then you are focused only on the sound in the the test lab.

If you have a sighted test using simple music of a speaker which you already own or are considering purchasing in a shop or at home, then you are not just influenced by the sound.

That does not mean you won't select the objectively better speaker, but it significantly reduces the odds of you getting the decision correct.