KEF R3 OR BOWER & WILKINS S706 s2 ???

[preload]

B&W use their speakers toed-in, in their own demo rooms. See attached images. Having owned several older B&W speakers through the years, I can't remember any advice against 0-deg / on-axis listening.

You can't make a general rule, since dispersion is all over the place with B&W these days. Some models have fairly even dispersion (but unfortunately uneven response to begin with), while others are horrible.

Sorry, but both of your photos show the B&W speakers only partially toed in and NOT aimed directly at the MLP. And that's exactly right. Perhaps you have misunderstood.

Whereas, a Genelec should be aimed directly at the MLP, and Harman blind listening tests were conducted in Monon with the MLP at 0 degs (ie speaker directly aimed at listener).

B&W's (having owned many different models) are only bright sounding if they are toed in too much. And if you look at the spins, and the on-axis response vs the much more smoothed off-axis response, it's clear that they are not intended to be listened to at 0-deg. I wonder how many folks make this mistake and assume the speakers are overly bright for this reason.

That might help. My own preference is toed-in, pointed at me, for the most direct sound. Plus, looking at that directivity data above, toed-out may not work well in any event.

That will make B&W's sound too bright, in my experience. The degree of toe-in affects the brightness of the speaker with 0-deg being very bright.

 

[Steve Dallas]

Sorry, but both of your photos show the B&W speakers only partially toed in and NOT aimed directly at the MLP. And that's exactly right. Perhaps you have misunderstood.

Whereas, a Genelec should be aimed directly at the MLP, and Harman blind listening tests were conducted in Monon with the MLP at 0 degs (ie speaker directly aimed at listener).

B&W's (having owned many different models) are only bright sounding if they are toed in too much. And if you look at the spins, and the on-axis response vs the much more smoothed off-axis response, it's clear that they are not intended to be listened to at 0-deg. I wonder how many folks make this mistake and assume the speakers are overly bright for this reason.

That will make B&W's sound too bright, in my experience. The degree of toe-in affects the brightness of the speaker with 0-deg being very bright.

Perhaps you missed my measurements on the previous page which show a pair of CM4s toed-in only about 15 degrees showing the classic Batman curve. Other B&W speakers post 2001 manufacture I have tested in-house have measured similarly, which caused me to abandon the brand. If you do not toe them in enough, the mids are greatly diminished due to design directivity errors, which are documented in this thread and others. And, toed in that much, they are much too bright.

Without a full spinorama, the listening window is a good baseline measurement. And, this measurement is used throughout this thread. According to Audioholics:

"A single measure of the responses that occur at and near the on-axis response is called the ‘listening window.’ The listening window curve is an average of horizontal responses at +/- 10, 20, and 30 degrees on the horizontal axis, and +/- 10 degrees on the vertical axis. It encompasses most of the area in which listeners would ever typically listen."

All measurements I have seen of recent B&W speakers have featured a listening window of which Batman would be proud. I hate that this is true as a former B&W fan. But it is true.

 

[Beave]

Sorry, but both of your photos show the B&W speakers only partially toed in and NOT aimed directly at the MLP. And that's exactly right. Perhaps you have misunderstood.

Whereas, a Genelec should be aimed directly at the MLP, and Harman blind listening tests were conducted in Monon with the MLP at 0 degs (ie speaker directly aimed at listener).

B&W's (having owned many different models) are only bright sounding if they are toed in too much. And if you look at the spins, and the on-axis response vs the much more smoothed off-axis response, it's clear that they are not intended to be listened to at 0-deg. I wonder how many folks make this mistake and assume the speakers are overly bright for this reason.

That will make B&W's sound too bright, in my experience. The degree of toe-in affects the brightness of the speaker with 0-deg being very bright.

Here are links to on- and off-axis measurements for several B&W speakers:

B&W 603
B&W 606
B&W S2
B&W 685 S2
B&W 704 S2
B&W 705 S2
B&W 705 Sig
B&W 805 D4

For every single one, you have to go *way* off-axis to get away from the rising treble response.

And for nearly every single one, when you do so, you get a huge hole in the low treble response.

 

[preload]

Perhaps you missed my measurements on the previous page which show a pair of CM4s toed-in only about 15 degrees showing the classic Batman curve. Other B&W speakers post 2001 manufacture I have tested in-house have measured similarly, which caused me to abandon the brand. If you do not toe them in enough, the mids are greatly diminished due to design directivity errors, which are documented in this thread and others. And, toed in that much, they are much too bright.

Without a full spinorama, the listening window is a good baseline measurement. And, this measurement is used throughout this thread. According to Audioholics:

"A single measure of the responses that occur at and near the on-axis response is called the ‘listening window.’ The listening window curve is an average of horizontal responses at +/- 10, 20, and 30 degrees on the horizontal axis, and +/- 10 degrees on the vertical axis. It encompasses most of the area in which listeners would ever typically listen."

All measurements I have seen of recent B&W speakers have featured a listening window of which Batman would be proud. I hate that this is true as a former B&W fan. But it is true.

The listening window is only valid if you're "listening" at 0-degs (directly on the tweeter axis).
This applies to speakers like the Genelec monitors:

And Revel loudspeakers:

Whereas, with B&W speakers, you're expected to "adjust" the amount of toe-in in order to adjust the relative brightness of the loudspeaker.
So, if you set up your B&W's in the same way as you would a Revel/Genelec, and you interpret the measurements as you would a Revel Genelec (i.e. assuming the "listening window" is usable when speakers are not aimed directly at you), then you will conclude that B&W's sound bright and the measurement prove it (which is just not the case, when set up properly).

Here are links to on- and off-axis measurements for several B&W speakers:

B&W 603
B&W 606
B&W S2
B&W 685 S2
B&W 704 S2
B&W 705 S2
B&W 705 Sig
B&W 805 D4

For every single one, you have to go *way* off-axis to get away from the rising treble response.

It just goes to show that you can't simply "eyeball" a series of polar FR measurements and predict the in-room response when the MLP is not at 0-degs with respect to the tweeter axis.

This is a MMM measurement of stereo pink noise of my B&W 802D's using a calibrated umik-1. As you can see, it follows a fairly smooth decreasing response with presence/BBC dip at 2-3kHz. This B&W set up in my living room sounds far better than the Genelec 8351B set up in my office. In fact, I use PEQ on my 8351B's to approximate the B&W curve! Sorry, but I prefer the way my music sounds on speakers with that dip, and so do a lot of people.

 

[Beave]

I see a pretty significant (~5dB) dip in your in-room response centered around 2.5kHz (going from 1.5kHz to 4kHz).

There is no axis for any of those B&W speakers that provides a good direct sound response or listening window response.

You can like that dip, and that's fine, but earlier you seemed to claim that one only needs to listen off-axis to get a good response from B&W speakers. Now you're re-defining "good" to suit your personal tastes.

 

[preload]

You can like that dip, and that's fine, but earlier you seemed to claim that one only needs to listen off-axis to get a good response from B&W speakers.

I believe you have misread. My specific comment was: "I wonder how many people who think B&W's sound bright are listening to them toed-in so that the listener is 0-deg on-axis (which is what you would do with a Genelec or a Revel)." And I think it's a fair question.

Now you're re-defining "good" to suit your personal tastes.

How a speaker actually sounds is more important than how it measures. Even using computerized analysis, Harman's own research was only able to show a partial correlation between a series of controlled measurements and listener preferences. And even then, the listener preferences were derived based on listening to a single mono loudspeaker ON-AXIS.

 

[Beave]

I believe you have misread. My specific comment was: "I wonder how many people who think B&W's sound bright are listening to them toed-in so that the listener is 0-deg on-axis (which is what you would do with a Genelec or a Revel)." And I think it's a fair question.{/quote]

No misreading. I answered that in order to reduce the brightness, one must listen at such an off-axis that there will be a large hole in the low treble. 15 deg doesn't seem to be enough for most of their models, nor even 30 deg or 45 deg for many of them. Do you really think people are listening to their B&W speaker at 50 or 60 degrees off axis to reduce the brightness?

How a speaker actually sounds is more important than how it measures. Even using computerized analysis, Harman's own research was only able to show a partial correlation between a series of controlled measurements and listener preferences. And even then, the listener preferences were derived based on listening to a single mono loudspeaker ON-AXIS.

How a speaker sounds correlates pretty well with how it measures, especially as far as tonality goes.

As I stated, there is no axis for any of those B&W speakers where one gets a good response on that axis. And the direct sound dominates what we hear.

You're doing all sorts of dancing, as usual, to justify your personal preference for your speakers. You like them, that's fine, but the attempts to justify their measured behavior are a reach.

 

[preload]

No misreading. I answered that in order to reduce the brightness, one must listen at such an off-axis that there will be a large hole in the low treble. 15 deg doesn't seem to be enough for most of their models, nor even 30 deg or 45 deg for many of them.

That's a pretty bold claim - you are essentially saying that you can look at a series of off-axis FR plots, rotate them 15/30/45-degs, average them out in your head, determine the predicted in-room response curve for each, and conclude that even 45-degs off-axis sounds bright? C'mon, please.

I just provided you with empiric evidence consisting of the ACTUAL in-room response curve of a pair of B&W 802D speakers. There is no "rising" treble.

And once again, that "presence dip" or "BBC dip" is 100% intentional.

How a speaker sounds correlates pretty well with how it measures, especially as far as tonality goes.

That's a pretty imprecise statement, mind you.
Consider the Harman research: Subjective loudspeaker preferences correlate SOMEWHAT with measurements, but only if those measurements are analyzed by a computer and are for loudspeakers that belong to the original sample playing a selection of rock music. Once you start talking about loudspeakers OUTSIDE that original sample playing non-rock music, then the correlation gets even worse. And THEN when folks start eyeballing off-axis measurements and think they can do better than the computerized analysis in terms of making predictions on subjective SQ, that's when I start rolling my eyes.
Now, tonality, I'll give that one to you - but only if you're looking at the in-room response curve, performing a linear regression, and measuring the slope of that linear regression line. And even if you DID do that, do you really think you could "eyeball" that slope with enough accuracy to determine if the speaker will sound bright, neutral, or dull? Negative, Captain.

You're doing all sorts of dancing, as usual, to justify your personal preference for your speakers. You like them, that's fine, but the attempts to justify their measured behavior are a reach.

A lot of people prefer the sound from B&W's, particularly the 800-series. It just so happens that quite a few folks on ASR prefer Revel/Genelec/etc. And that's fine too.

And personally, I'm not obsessed with finding loudspeakers that fit a specific measurement paradigm of "perfection," particularly since that paradigm has not been demonstrated to have a perfect correlation with blind listening preferences. If a speaker sounds better to me, it sounds better to me, and if the measurements don't explain it, then perhaps it's time to revisit how we're interpreting those measurements. That's the difference between experimental science and engineering.

 

[Beave]

That's a pretty bold claim - you are essentially saying that you can look at a series of off-axis FR plots, rotate them 15/30/45-degs, average them out in your head, determine the predicted in-room response curve for each, and conclude that even 45-degs off-axis sounds bright? C'mon, please

I made no such claim. I simply pointed out that the DIRECT SOUND will not be objectively good (flat) at any axis. I said nothing about the speaker sounding "bright" at extreme off-axis angles. In fact, quite the opposite, I said that it will not be bright at extreme off-axis angles but will have a hole in the lower treble.

If you listen to them 180 deg off axis, they will not be bright. I agree with you there. Are you happy now? Are you ready to stop the sophistry?

I just provided you with empiric evidence consisting of the ACTUAL in-room response curve of a pair of B&W 802D speakers. There is no "rising" treble.

And once again, that "presence dip" or "BBC dip" is 100% intentional.

Not in the in-room response. So what? Can you point to research showing that the in-room response takes precedence over the direct anechoic response of a speaker in terms of listener preference?

Maybe that dip is intentional. You like it. I don't. So what, again.

And personally, I'm not obsessed with finding loudspeakers that fit a specific measurement paradigm of "perfection," particularly since that paradigm has not been demonstrated to have a perfect correlation with blind listening preferences. If a speaker sounds better to me, it sounds better to me, and if the measurements don't explain it, then perhaps it's time to revisit how we're interpreting those measurements. That's the difference between experimental science and engineering.

Once again you seem to think that your personal, sighted preferences somehow matter to others or somehow throw doubt on the established body of research. They don't.

A lot of people prefer the sound from B&W's, particularly the 800-series. It just so happens that quite a few folks on ASR prefer Revel/Genelec/etc. And that's fine too.

They prefer the sound from B&Ws over what? In level matched blind tests or in sighted evaluations?

 

[thewas]

I just provided you with empiric evidence consisting of the ACTUAL in-room response curve of a pair of B&W 802D speakers. There is no "rising" treble.

As Beave correctly wrote the steady state response doesn't tell much about the perception of the upper frequency band as there the direct sound is dominating more. Also, since directivity is usually increasing at the upper octaves as well as room absorption the steady state response can be even falling with a rising direct sound.

John Atkinson provides often also such steady state in room measurements of the loudspeakers he measures and the elevated treble is obvious when compared with anechoically more neutral ones, for example:

Fig.7 Bowers & Wilkins 804 D4, spatially averaged, 1/6-octave response in JA's listening room (red) and of the PSB Synchrony T600 (blue).

The Bowers & Wilkins 804 D4's measured performance is enigmatic, with positive aspects—the relatively high sensitivity, the superbly quiet enclosure, the well-controlled dispersion, and the excellent pair matching—that must be set against others, like the demanding impedance, the excessive high frequencies, and the resonances in the port's and upper-frequency units' output.

Source: https://www.stereophile.com/content/bowers-wilkins-diamond-series-804-d4-loudspeaker-measurements

A lot of people prefer the sound from B&W's, particularly the 800-series.

Dē gustibus nōn est disputandum, a lot of audiophiles also supposedly prefer vinyl sound, highly distorting tube amps and full range drivers, possibly also because they are used to their sound, haven't seen a controlled blind listening test though with such an outcome yet.

By the way Dali tunes their loudspeakers to be not toed in, but they also state it in their manuals and also measure quite flat around 15°-30°.

 

[TrevC]

B&W use their speakers toed-in, in their own demo rooms. See attached images. Having owned several older B&W speakers through the years, I can't remember any advice against 0-deg / on-axis listening.

You can't make a general rule, since dispersion is all over the place with B&W these days. Some models have fairly even dispersion (but unfortunately uneven response to begin with), while others are horrible.

See red trace. Pretty flat. The green is the bass distorting and somewhat overrated LS50, the red is Marten Oscar Duo.

 

[Beave]

See red trace. Pretty flat. The green is the bass distorting and somewhat overrated LS50, the red is Marten Oscar Duo.

That's the 705 Signature. Measurements here: 705 Sig from Stereophile

Yes, it looks pretty flat. That's the in-room response. Who wants a flat in-room response? I would guess most people would prefer a down-sloping in-room response.

 

[TrevC]

That's the 705 Signature. Measurements here: 705 Sig from Stereophile

Yes, it looks pretty flat. That's the in-room response. Who wants a flat in-room response? I would guess most people would prefer a down-sloping in-room response.

Surely we are talking measurements here, not what some people prefer to listen to.

 

[thewas]

Surely we are talking measurements here, not what people prefer to listen to.

A neutral loudspeaker has flat direct sound, not steady state sound at the listening position, which is in typical living rooms and with typical increasing directivity loudspeakers decreasing to the higher frequencies for such neutral loudspeakers.

 

[Ageve]

Sorry, but both of your photos show the B&W speakers only partially toed in and NOT aimed directly at the MLP. And that's exactly right. Perhaps you have misunderstood.

No, why are you being dishonest?

They are clearly toed-in, with 0 deg towards the listening position. And it's in B&Ws own demo room.

See red trace. Pretty flat. The green is the bass distorting and somewhat overrated LS50, the red is Marten Oscar Duo.

No, it's not "pretty flat".

From 2 kHz to 8 kHz, there's a change of at least 3 dB, in the wrong direction. It should be down-sloping in room.

Surely we are talking measurements here, not what people prefer to listen to.

No, it's the exact opposite. People (without bias) prefer a down-sloping treble in room.

The Subjective and Objective Evaluation of Room Correction Products

In a recent article, I discussed audio’s circle of confusion that exists within the audio industry due to the lack of performance standards...

seanolive.blogspot.com

A flat in-room target response is clearly not the optimal target curve for room equalization. The preferred room corrections have a target response that has a smooth downward slope with increasing frequency. This tells us that listeners prefer a certain amount of natural room gain. Removing the room gain, makes the reproduced music sound unnatural, and too thin, according to these listeners. This also makes perfect sense since the recording was likely mixed in room where the room gain was also not removed; therefore, to remove it from the consumers' listening room would destroy spectral balance of the music as intended by the artist.

 

[TrevC]

Yet more 'my way or the highway' subjective opinions. You see, I too know what I like to listen to and what I don't, and I don't like small speakers with distorted bass and rolled off highs but I love the 706 S2 sound on all types of music. A grand well spent on the best speakers I've ever owned.
I don't care what Floyd Toole thinks either.

 

[Digital_Thor]

Surely we are talking measurements here, not what some people prefer to listen to.

You have to distinguish between short term wow-effect and long term enjoyment. And I believe that no matter your preference... you would enjoy the more neutral speaker - for longer - meaning better investment and hugely better enjoyment for the buyer.
I measured several B&W speakers, from the cheaper models to the 800 series. And surely they have that horrid frequency response, that simply screams bad design - all of them. I believed that the good designer of the older B&W speakers - prior to ca. 2000 - ran off and helped Vivid Audio.
Besides - a speaker that has a difference of almost 8dB from midrange to tweeter - in the most sensitive range of our hearing - just silly!

I don't care how "famous" they are. We no longer have to be blind and ignorant of such poorly made speakers - no matter the fancy exterior... or interior for that matter.
It's carved in granite, that they made a hugely "colored" sounding speaker, with a direct focus on anything else but accurate rendition of the original input signal.
So - if you ask for accurate - clearly go somewhere else. If you just want eye candy ( taste may differ ) and have loads of money to burn, then be my guest.

And yes - I did hear them all, and I instantly recognized the sound signature when I entered a room - may it be at a private home or huge exhibition. I always think - " hey mate - did you burn your midrange?"

No matter the room, the volume, the music, the people etc. Compared to B&W - when I hear a KEF, JM/Focal, Revel and the like. I don't have to sit perfectly or listen to a special type of music... it's just on average way more enjoyable with the technically superior speaker designs.

And I primarily focus on the top octaves in this argument - since the room has way more influence below around 500Hz.

A person ask for advice about speakers, and openly writes that he is not the most experienced within the field. Then I believe it would be the kindest, to not point him to very expensive options, that also technically are very inferior to most other speakers on the marked - don't you think?

And, as pointed out clearly a few times in this thread. A well-designed loudspeaker can pretty easily be EQ'ed to one's lickings. B&W is severely flawed to begin with - and therefore EQ will just ruin it even more - pushing you further away from the original source signal.

You can only successfully EQ the frequency response of any speaker, if the same bump or dip is represented at all acoustical angles in front of the speaker.

If not drivers, filters, cabinet shape, cross-over, placement of drivers all match up to one smooth sum, creating that nice even power response all over. Then any EQ will be less effective or even just move an issue around - never really solving it.

My bet - in a proper blind test - any well-designed speaker from about 1000$ and up, would beat any B&W speaker - at any price, from the last 20 years.

 

[thewas]

Yet more 'my way or the highway' subjective opinions. You see, I too know what I like to listen to and what I don't, and I don't like small speakers with distorted bass and rolled off highs but I love the 706 S2 sound on all types of music. A grand well spent on the best speakers I've ever owned.
I don't care what Floyd Toole thinks either.

The subjective opinion is exactly what you write, not what Toole writes. Also no one discusses here about rolled off highs (you still seem to confuse flat direct sound with flat steady state measured at the LP) or distorted bass (it is funny though if you think that your 706 S2 are not small speakers with distorted bass like almost all typical passive compact 2-way loudspeakers).

 

[Ageve]

Yet more 'my way or the highway' subjective opinions. You see, I too know what I like to listen to and what I don't, and I don't like small speakers with distorted bass and rolled off highs but I love the 706 S2 sound on all types of music. A grand well spent on the best speakers I've ever owned.

So, you like your speakers? Then enjoy them, but don't try to convince others that they are flawless, based on your subjective opinions, and by ignoring facts.

I don't care what Floyd Toole thinks either.

Did you read the article?

It wasn't about "what Floyd Toole thinks". It was based on a listening test by Sean Olive, where people preferred a down-sloping treble when using room correction. Sean Olive only explained the reason why it sounds better.

believed that the good designer of the older B&W speakers - prior to ca. 2000 - ran off and helped Vivid Audio.

That's correct. Laurence Dickie used to work for B&W. He left after designing the 800 Nautlius series (and of course the Nautlius speaker). I'm not sure, but I think the 800 Nautlius series was released in 1998.

 

[TrevC]

My bet - in a proper blind test - any well-designed speaker from about 1000$ and up, would beat any B&W speaker - at any price, from the last 20 years.

Wouldn't you describe the LS50 as well designed? Well I'm afraid it doesn't come anywhere near to beating the 706 at the same price.