KEF R3 OR BOWER & WILKINS S706 s2 ???

[MaxBuck]

Good grief. B&W speakers have a house sound; no one with any sense denies that. KEFs not so much. It's unlikely anyone would be equally happy with both.

Also, "preference graphs" are meaningless for any individual listener.

Buy whatever you prefer, but for heaven's sake quit nattering on about this stuff.

ETA: apparently I'm in a foul mood this evening. Probably best to ignore me.

 

[MAB]

Good grief. B&W speakers have a house sound; no one with any sense denies that. KEFs not so much. It's unlikely anyone would be equally happy with both.

Also, "preference graphs" are meaningless for any individual listener.

Buy whatever you prefer, but for heaven's sake quit nattering on about this stuff.

ETA: apparently I'm in a foul mood this evening. Probably best to ignore me.

You are so right. If people like B&W house sound, that's great. I'm pretty sure my tastes aren't relevant to others. This has turned into a Monty Python skit.

 

[MAB]

In case the original topic is still relevant, I would get the R3.
In either case, you should plan on getting a pair of subwoofers if space and funds permit.

 

[wjp007]

Good grief. B&W speakers have a house sound; no one with any sense denies that. KEFs not so much. It's unlikely anyone would be equally happy with both.

Also, "preference graphs" are meaningless for any individual listener.

Buy whatever you prefer, but for heaven's sake quit nattering on about this stuff.

ETA: apparently I'm in a foul mood this evening. Probably best to ignore me.

Well stated. Sound preference is a personal thing. There is no right or wrong, it's just what you preferred. Having just switched from LS50 Meta (which many KEF people feel are too bright) to 705 s2, I prefer the 705 s2. I wouldn't have switched if it wasn't for the deal I got on open box units. IMHO both speakers are fantastic and I would be happy with either speaker.

I am interested to understand how pointing the speakers directly at you improves stereo imaging. I've always thought the key to great imaging is to reduce first reflections. Does pointing the speakers right at you, reduce the amount of sound that hits the first reflection as the tweeters will be significantly off axis. In my room this would be like 60 degrees off center.

 

[Steve Dallas]

I never really liked the lower-end B&W models either, including the CM (which I believe was somewhere between a 600 and 700 series in their product line). At the same time, I don't believe the CM-series was really designed to represent the hallmark of B&W engineering, I considered it "there" simply to fill a certain market segment.


Ok, that explains it, thanks.


I would suggest that the BBC dip is intentional and shouldn't be EQ'd for that reason.
I suspect that the BBC dip followed by a low Q hump in the upper treble is actually what makes the higher end B&W’s sound good to many people, particularly if toe-in and “room furnishings” are used to tweak the level of brightness to taste.

The CM series replaced the 700 series for a time. Not exactly lower end. From the Absolute Sound:

Sitting pretty between the studio-quality speakers of the Bowers & Wilkins 800 Series Diamond and the surprisingly affordable new 600 Series, the new CM Series delivers stunning audiophile sound quality and subtle, stylish design perfect for home audio. The new CM Series truly is beautiful sound.

The BBC dip had a very narrow and specific purpose: to reduce a cone resonance of the drivers used at the time. A side effect was changing the soundstage depth for the engineers using them nearfield in small control rooms and vans. Sources: Harbeth and Dick Gundry's son.

B&W's dip is much more than the BBC dip. It is often up to -6dB in some B&W speakers (-5dB in yours, -6dB in mine). Whether to EQ the BBC dip is a matter of personal opinion. I hate having that frequency range sucked out and absolutely want to EQ it flat. We have already shown that toeing speakers out helps the treble peak, but worsens the dip, which is why B&W speakers are no longer for me.

 

[MAB]

The CM series replaced the 700 series for a time. Not exactly lower end. From the Absolute Sound:



The BBC dip had a very narrow and specific purpose: to reduce a cone resonance of the drivers used at the time. A side effect was changing the soundstage depth for the engineers using them nearfield in small control rooms and vans. Sources: Harbeth and Dick Gundry's son.

B&W's dip is much more than the BBC dip. It is often up to -6dB in some B&W speakers (-5dB in yours). Whether to EQ the BBC dip is a matter of personal opinion. I hate having that frequency range sucked out and absolutely want to EQ it flat. We have already shown that toeing speakers out helps the treble peak, but worsens the dip, which is why B&W speakers are no longer for me.

Nice summary of the B&W house sound, and the interaction of on and off-axis behavior.

 

[preload]

The CM series replaced the 700 series for a time. Not exactly lower end. From the Absolute Sound:

I know it's not critically important here, but I believe the 700 series was the replacement (an improvement) over the CM series:

B&W Launches the new 700 Series: replacing the CM Series

The new range include six different 2-channel designs The new speakers include: the 702 S2 (£3,299), featuring a carbon dome tweeter in a solid-body housing and a decoupled Continuum midrange, this is the 700 Series’ flagship floorstander. 705 S2 (£1,799), also features the ‘tweeter-on-top’...

theaudiophileman.com

Regardless, neither the discontinued CM series of the 700 series are really the same as the 800-series. They serve a different segment of the market (with perhaps a bit of cross-shopping).

The BBC dip had a very narrow and specific purpose: to reduce a cone resonance of the drivers used at the time. A side effect was changing the soundstage depth for the engineers using them nearfield in small control rooms and vans. Sources: Harbeth and Dick Gundry's son.

I've heard a lot of different rationales for the original BBC dip and other treble dips in that same region of various shapes and magnitudes. I also don't think that engineers are stupid in the sense that when the original rationale (for instance, cone resonance) was solved, that they decided to leave the dip in anyway, even though there was no longer a reason to and it would make the speakers sound worse.

Also, the soundstage on my 802D's is absolutely magnificent - it is large, 3-d, and glorious, and I've literally turned my head to the side because I thought an instrument physically appeared in my room. I realize the room is different, but I've never experienced that sort of imaging from the Genelecs, not even close in terms of size, depth, or anything.

B&W's dip is much more than the BBC dip. It is often up to -6dB in some B&W speakers (-5dB in yours, -6dB in mine). Whether to EQ the BBC dip is a matter of personal opinion. I hate having that frequency range sucked out and absolutely want to EQ it flat. We have already shown that toeing speakers out helps the treble peak, but worsens the dip, which is why B&W speakers are no longer for me.

Perhaps in isolation, the "B&W dip" (let's just call it that) isn't desirable. However, when coupled with the gentle upper-treble hump, it appears the combination works. Adjusting the toe simply adjusts the amount of that upper-treble hump. It's also very possible that the combination of the B&W dip and the B&W hump is strongly preferred by a subset of listeners. And the fact that people seem to either really love or really hate B&W's is consistent with this.

 

[Steve Dallas]

I know it's not critically important here, but I believe the 700 series was the replacement (an improvement) over the CM series:

B&W Launches the new 700 Series: replacing the CM Series

The new range include six different 2-channel designs The new speakers include: the 702 S2 (£3,299), featuring a carbon dome tweeter in a solid-body housing and a decoupled Continuum midrange, this is the 700 Series’ flagship floorstander. 705 S2 (£1,799), also features the ‘tweeter-on-top’...

theaudiophileman.com

Regardless, neither the discontinued CM series of the 700 series are really the same as the 800-series. They serve a different segment of the market (with perhaps a bit of cross-shopping).


I've heard a lot of different rationales for the original BBC dip and other treble dips in that same region of various shapes and magnitudes. I also don't think that engineers are stupid in the sense that when the original rationale (for instance, cone resonance) was solved, that they decided to leave the dip in anyway, even though there was no longer a reason to and it would make the speakers sound worse.

Also, the soundstage on my 802D's is absolutely magnificent - it is large, 3-d, and glorious, and I've literally turned my head to the side because I thought an instrument physically appeared in my room. I realize the room is different, but I've never experienced that sort of imaging from the Genelecs, not even close in terms of size, depth, or anything.


Perhaps in isolation, the "B&W dip" (let's just call it that) isn't desirable. However, when coupled with the gentle upper-treble hump, it appears the combination works. Adjusting the toe simply adjusts the amount of that upper-treble hump. It's also very possible that the combination of the B&W dip and the B&W hump is strongly preferred by a subset of listeners. And the fact that people seem to either really love or really hate B&W's is consistent with this.

The CM series replaced the 700 series circa 2000 and was then replaced by a reintroduced 700 series in 2017. However the lineage is difficult to nail down in publications. Some sources say the CM series was born in 2014, however I purchased mine in 2001, and found a review of them written in 2002. Then there is this curious timeline published in sister magazine Stereophile, which places the CM series from 2008 through 2017. That may be referring to the Chinese manufacture version. Mine were made in the UK. Which is probably why they leaked oil and eventually broke down.

I happen to know for a fact the 700 series existed before the CM series, because I owned a pair of 700 series bookshelves before I upgraded to my 804N and 805N. I asked the dealer (Ben at Home Theater Shack in Southlake, Texas, now Modia in Plano, Texas) about the midsize 700 series tower, and he told me the 700 series was now the CM series, and B&W was marketing them toward the home theater application. I said, "Whatever, throw in a pair of those," and went home with 3 pairs of speakers. The CM series persisted through a few incarnations until changing back to the 700 series in 2017.

Soundstage is impossible to debate on the internet, and for every one of you, there will be a Genelec fan who claims the opposite. It is also highly room and setup dependent. I can make the same claim about my Revel F206s which replaced my 804Ns. The soundstage and imaging are absolutely glorious. The KEF R3s which replaced my 805Ns are more about point source imaging, but it has to be that way in the problematic room they inhabit and were chosen for the application. The 805s sounded much too diffuse in that room.

I think we can agree on the B&W dip. You are in the subset of those who like it. I am most definitely not in favor missing 4000Hz of midrange information.

BUT... I'll take this opportunity for a little experiment. I'll create a curve in Dirac with a BBC dip for the R3s and report back on what I think about it. (I can do this, because the R3s have no directivity error in that range.)

 

[preload]

The CM series replaced the 700 series circa 2000 and was then replaced by a reintroduced 700 series in 2017. However the lineage is difficult to nail down in publications. Some sources say the CM series was born in 2014, however I purchased mine in 2001, and found a review of them written in 2002. Then there is this curious timeline published in sister magazine Stereophile, which places the CM series from 2008 through 2017. That may be referring to the Chinese manufacture version. Mine were made in the UK. Which is probably why they leaked oil and eventually broke down.

I happen to know for a fact the 700 series existed before the CM series, because I owned a pair of 700 series bookshelves before I upgraded to my 804N and 805N. I asked the dealer (Ben at Home Theater Shack in Southlake, Texas, now Modia in Plano, Texas) about the midsize 700 series tower, and he told me the 700 series was now the CM series, and B&W was marketing them toward the home theater application. I said, "Whatever, throw in a pair of those," and went home with 3 pairs of speakers. The CM series persisted through a few incarnations until changing back to the 700 series in 2017.

You could be right about that the history, it does sound familiar. I also remember there was a CDM series that existed around 2000 (I owned a pair back then). And of course, it's all marketing anyway - B&W can assign whatever name they want to each production line of speakers, though for product differentiation, we know that the 800-series tends to cost more and is their flagship line.

Soundstage is impossible to debate on the internet, and for every one of you, there will be a Genelec fan who claims the opposite. It is also highly room and setup dependent. I can make the same claim about my Revel F206s which replaced my 804Ns. The soundstage and imaging are absolutely glorious.

I actually haven't read great things about Genelec soundstage, perhaps because the recommendation placement is often right up against the backwall. But I agree with you that the soundstage is room/placement dependent. The point I wanted to make is that the presence of a "B&W" dip doesn't necessarily exclude the possibility of an exceptional soundstage.

BUT... I'll take this opportunity for a little experiment. I'll create a curve in Dirac with a BBC dip for the R3s and report back on what I think about it. (I can do this, because the R3s have no directivity error in that range.)

Sounds like a fun experiment. If possible, I would suggest shaping the dip in DIRAC to match that of the B&W curve I posted above. I found that when EQ'ing my Genelecs, when I was "close enough" on the curve, it still didn't sound quite right. And don't forget the upper treble hump after the dip.

 

[Ageve]

706 S2 measurements are available here:

BOWERS & WILKINS 706 S2

Seria 700 S2 jest środkową w ofercie Bowersa, a model 706 S2 środkowym wśród podstawkowych modeli tej serii. Bowers nie pożałował – ...

audio.com.pl

As expected, the tweeter is way too bright, but there are other problems as well.

Please note: I don't speak Polish. It's all Google Translate.

Red: On-axis
Dark red: 15 deg off-axis (horizontal)
Yellow: 30 deg off-axis (horizontal)
Blue: 7 deg above (górę = up) **
Green: 7 deg below *
Black: With grille on

From the review:

The processing characteristics of the 706 S2 do not look perfect, but familiar - the distribution of unevenness is not accidental, but characteristic and resulting from specific solutions adopted many years ago by B&W designers - although not by John Bowers himself...

In his time, the characteristics were better aligned, but at the beginning of the 20th century there was a shift in priorities. Gentle filtering at a high crossover frequency is not the best way to achieve a smooth transition, let alone stabilize it on different axes.
...
Here we see the resultant characteristics, measured on different axes; not the worst one appears at an angle of +7 deg ** (from a distance of more than a meter we have to see the top).

It is worth considering that in such a situation we increase the difference in distance from the listening (measuring) place to the acoustic centers of both transducers (we approach the tweeter) - contrary to the concept of "time alignment".

In order to obtain at least a relatively well-balanced response, the priority is the phase alignment of the loudspeakers in the crossover frequency range, which, with the shifts introduced by the drivers and filters themselves, does not have to be associated with time alignment (or the same polarization of both drivers - in this case, however, consistent).

At the -7º angle*, the attenuation at 2.5 kHz is the greatest (which is not due to the directional characteristics of the individual drivers, but the greatest phase shift between them), although you can see a higher level in the 4-7 kHz range, where the speakers are best matched at this angle (and least at +7º **), the relative phase shift changes as a function of frequency.

The "intermediate" characteristic - with slight attenuations at 2.5 kHz and 6 kHz - is obtained on the main axis. The manufacturer promises to maintain the characteristics in the +/-3 dB path in the 50 Hz - 28 kHz band, and it is possible, but on the +7º axis. **

+7 deg vertical: Reasonably flat up to 3 kHz.
0 deg vertical: 6 dB dip at 2 kHz.
-7 deg vertical: 11-12 dB dip at 2.5 kHz.

The 2-3 kHz dip is caused by poor crossover design, just like with the 607 S2:

Bowers & Wilkins 607 S2 Anniversary Edition Review

This is a review and detailed measurements of the B&W 607 S2 Anniversary Edition. It was kindly sent to me by a member and costs US $700 through their various dealers. The 607 S2 comes in different color. My same was in black: Disappointing to see no attempt at directivity control with a...

www.audiosciencereview.com

Based on these measurements, the 706 S2 is almost as bad as the 607 S2.

edit: KEF R3:

KEF R3

Konstrukcja podstawkowa z układem trójdrożnym to rzadkość. Od kilu lat widzimy nostalgiczną modę na paczki w starym stylu – dość duże, ...

audio.com.pl

(Black: grille on)

 

[preload]

"In his time, the characteristics were better aligned, but at the beginning of the 20th century there was a shift in priorities"

Wow B&W had pretty good technology for the 1900's.

 

[Ageve]

Wow B&W had pretty good technology for the 1900's.

John Bowers was designing accurate loudspeakers in the late 1800's. That's even more impressive.

Google Translate at its finest.

 

[Steve Dallas]

This is an interesting read regarding the evolution of B&W's 800 series as perceived by these fellows:

Bowers & Wilkins 803 D4 Loudspeakers: A Secrets Video Review - HomeTheaterHifi.com

Having previously reviewed the 803 D3 back in 2017, Bowers and Wilkins were kind enough to extend the invitation for me to review the...

hometheaterhifi.com

Also, I did perform that BBC dip testing I mentioned. The results are... interesting. I'll write them up when I have time. Yes, I have measurements.

 

[Ageve]

This is an interesting read regarding the evolution of B&W's 800 series as perceived by these fellows:

Interesting measurements as well.

It's not possible to tame the tweeter by moving off-axis (horizontal). Perhaps not surprising since the 805 D4 behaves in the same way.

SoundStageNetwork.com | SoundStage.com - NRC Measurements: Bowers & Wilkins 805 D4 Loudspeakers

www.soundstagenetwork.com

The dip gets worse when moving down from the tweeter axis, just like with the 706 S2, but it also moves up in frequency:

"Going down things get much worse. One could point to the good news that one likely listens seated at a higher position, but the wider negative angles are coming back to the ear from the floor bounce.

Note how the maxima of the dip moves up with an increasing negative angle. We would not expect this if the notch was due to midrange–tweeter physical displacement resulting in a cancelation at the crossover point.

I can come up with only one reason; the speaker has evolved from its 1980 performance to what we have here. More people purchase this version than the original."

The old Matrix 801 and S2 reviews were interesting as well.

https://worldradiohistory.com/Archive-All-Audio/Archive-High-Fidelity/80s/High-Fidelity-1980-10.pdf

"In simplest possible terms, the 801 is among the handful of great loudspeakers available. Detail, whether of timbre or of stereo imaging, is reproduced with a
delicate accuracy that imposes no obtruding quality of its own. That very unobtrusiveness is a mark of the design's success, though it puts it at a disadvantage
in hasty A/B comparisons, where less accurate speakers may seem more dramatic. Indeed, with extended listening, the little exaggerations on which their appeal is based become progressively apparent, while the 801's superb poise and balance make it seem, more and more, the objective reference against which the others' foibles can be measured."

https://worldradiohistory.com/Archive-All-Audio/Archive-Audio/90s/Audio-1990-11.pdf

 

[preload]

Interesting measurements as well.

It's not possible to tame the tweeter by moving off-axis (horizontal). Perhaps not surprising since the 805 D4 behaves in the same way.

SoundStageNetwork.com | SoundStage.com - NRC Measurements: Bowers & Wilkins 805 D4 Loudspeakers

www.soundstagenetwork.com

What on earth are you talking about now?
The soundstage measurements (your link) clearly shows a decreasing treble response of the 805D4 as you move further off-axis horizontally. Are you sure you know what you're looking at?

 

[preload]

This is an interesting read regarding the evolution of B&W's 800 series as perceived by these fellows:

Bowers & Wilkins 803 D4 Loudspeakers: A Secrets Video Review - HomeTheaterHifi.com

Having previously reviewed the 803 D3 back in 2017, Bowers and Wilkins were kind enough to extend the invitation for me to review the...

hometheaterhifi.com

Also, I did perform that BBC dip testing I mentioned. The results are... interesting. I'll write them up when I have time. Yes, I have measurements.

Thanks for sharing this link, @Steve Dallas. I hadn't seen it before. The analysis of the treble dip followed by slight treble boost is consistent with my own measurements and experience.

I didn't watch the included video review in its entirety, but from the clips I did see, it was interesting that:
1) The reviewer tried the speakers in 2 different rooms and had 2 completely different impressions - I've also had my 802D's in 2 different living rooms, and I can say I had a similar experience. In the smaller room, where I needed to keep the 802D's up against the wall, the bass was great, but the soundstage suffered in terms of depth. In the larger room, where the 802D's were a few feet away from the back wall, the bass suffered, but the soundstage was incredibly 3-d. I agree that the SQ is highly placement/room dependent. But not big deal on the bass, these speakers need a subwoofer either way.
2) The reviewer compared his experience with the 803D4 and his recollection of the Revel F328Be and noted that the character was different, but one was not necessarily better. I found that interesting.
3) The reviewer chose to position the speakers only slightly toed-in (i.e. not angled in to aim directly at the MLP), which supports what I was suggesting earlier.

 

[Ageve]

What on earth are you talking about now?
The soundstage measurements (your link) clearly shows a decreasing treble response of the 805D4 as you move further off-axis horizontally. Are you sure you know what you're looking at?

Did I say anything about the tweeter level staying exactly the same off-axis?

The correct answer is: No.

This is what I said:

It's not possible to tame the tweeter by moving off-axis (horizontal). Perhaps not surprising since the 805 D4 behaves in the same way.

At 30 deg, the (804 D4) response is basically the same as on-axis, down to about 1.5 kHz. The tweeter level is reduced by 2-3 dB, but so is everything else above 1.5 kHz.

The reviewer chose to position the speakers only slightly toed-in (i.e. not angled in to aim directly at the MLP), which supports what I was suggesting earlier.

Based on the 805 D4 measurements (same story with 804 D4, but it's easier to illustrate with the Soundstage images):
A slight angle (15 deg) will reduce the tweeter level by a single dB.
A not so slight angle (30 deg) will reduce the tweeter level by just 2 dB. It will also make the 3 kHz dip even worse.

Yep. Problem solved. Nothing to see here.

Difference between 500 Hz and 8-9 kHz:
0 deg: 8 dB
15 deg: 7 dB
30 deg: 6 dB

Difference between 1.5 kHz and 8-9 kHz:
0 deg: 8 dB
15 deg: 8 dB
30 deg: 7 dB

Difference between the 3 kHz dip and 8-9 kHz:
0 deg: 9 dB
15 deg: 9 dB
30 deg: 10 dB

I'm really trying to understand your reasoning. Are you using the peak at 100 Hz as a reference, or perhaps the resonances (Batman ears)?

I mean, even at 45 deg, the tweeter is 5 dB higher than the 500 Hz level. You would need a 60 deg angle to avoid it. When using slight toe-in as you recommend (15 deg in the image), it would look like this:

Blue: 500 Hz (below the first Batman ear)
Green: 45 deg
Red: 60 deg

A few quotes from the video review (a positive review, I know):

If anything, I think the speakers, because of their... [video cut] extra bit going on with the treble, or the air, I guess, with the detail and stuff... On occasion it made me want to say that, maybe the midrange is a little bit reserved, but... I don't really think so. I think the midrange on these speakers is nicely balanced. I think that... all that extra detail, and... you know, extra energy I'm hearing in the treble region is... possibly making me feel like it's... the midrange is not quite in balance, but I'm sure it is.

I think what we have with these speakers is... an extra oomph with the... eh... with the treble and the air and the detail and stuff.
...
It will reveal every single flaw... times ten. These will make a bad recording sound worse. So... Feed them good stuff.
...
One other thing you might want to consider, is that if you are a avid vinyl listener, you may want to take a close look at the pairing of your phono cartridge, that you're using with these speakers. I tested a number of cartridges that I had on hand, with... on my turntables with these speakers... And I tended to notice that cartridges that had a... it's sort of... they were voiced with a tipped up high end, or slightly more exaggerated high end... I found to be... not a very good match with these speakers...

They... It just made the high end... the upper mid and the highs like *really*... a little too much. Things got very edgy and slightly abrasive [video cut] whereas when I paired them with cartridges that were you know, more linear response... it sounded much better.
...
As... razor sharp and... slightly unforgiving as these speakers are, I would definitely keep into consideration your... the type of phono cartridge you are using.
...
The treble response is... extended. Very energetic, and very, very detailed [video cut]. In fact, the extended detail and precision of the sound of these speakers, pretty much brings up my only potential criticism. If you are at all sensitive to... the extended treble energy that the 804 D4 put out, you know they might not necessarily, you know be the speakers for you.

They are incredibly detailed and revealing to the point of being borderline ruthless. They... will not do poorly recorded music any favours, so... Am... Unless you feed them well recorded material on a consistent basis, do not expect them to smooth over, you know, any rough edges of... some of your lower quality recordings. They will not be merciful to any degree on that regard.

beat around the bush - Wiktionary, the free dictionary

en.wiktionary.org

Notice how he is referring to the treble as being "extended", "very energetic", having "extra oomph", "extra detail" and "extra bit going on".

Of course this "extended detail" will be impressive at first. That's probably why they design them this way.

From the old 801 Matrix review again:

"In simplest possible terms, the 801 is among the handful of great loudspeakers available. Detail, whether of timbre or of stereo imaging, is reproduced with a
delicate accuracy that imposes no obtruding quality of its own. That very unobtrusiveness is a mark of the design's success, though it puts it at a disadvantage
in hasty A/B comparisons, where less accurate speakers may seem more dramatic. Indeed, with extended listening, the little exaggerations on which their appeal is based become progressively apparent, while the 801's superb poise and balance make it seem, more and more, the objective reference against which the others' foibles can be measured."

 

[preload]

Did I say anything about the tweeter level staying exactly the same off-axis?

The correct answer is: No.

This is what I said:



At 30 deg, the (804 D4) response is basically the same as on-axis, down to about 1.5 kHz. The tweeter level is reduced by 2-3 dB, but so is everything else above 1.5 kHz.



Based on the 805 D4 measurements (same story with 804 D4, but it's easier to illustrate with the Soundstage images):
A slight angle (15 deg) will reduce the tweeter level by a single dB.
A not so slight angle (30 deg) will reduce the tweeter level by just 2 dB. It will also make the 3 kHz dip even worse.

Yep. Problem solved. Nothing to see here.

View attachment 251488

Difference between 500 Hz and 8-9 kHz:
0 deg: 8 dB
15 deg: 7 dB
30 deg: 6 dB

Difference between 1.5 kHz and 8-9 kHz:
0 deg: 8 dB
15 deg: 8 dB
30 deg: 7 dB

Difference between the 3 kHz dip and 8-9 kHz:
0 deg: 9 dB
15 deg: 9 dB
30 deg: 10 dB

I'm really trying to understand your reasoning. Are you using the peak at 100 Hz as a reference, or perhaps the resonances (Batman ears)?

I mean, even at 45 deg, the tweeter is 5 dB higher than the 500 Hz level. You would need a 60 deg angle to avoid it. When using slight toe-in as you recommend (15 deg in the image), it would look like this:
View attachment 251470

Blue: 500 Hz (below the first Batman ear)
Green: 45 deg
Red: 60 deg
View attachment 251497

OH WOW, you honestly do NOT know how to interpret off-axis response charts!!!!!! You really honestly don't know, yet you're pretending to analyze charts like you do.
Put simply, treble balance can be determined by the slope of the regression line through the FR curve. It should be plainly obvious that the slope is decreasing as you head more and more off-axis on the 805D4 charts you linked and referred to. Even people who disagree with my perspective on how B&W speakers sound can plainly see this in the charts.

But even "eyeballing it," where you would typically use 100-200hz as your "anchor," it's still plainly obvious that the AUC of the treble region is falling with increasing off-axis response. It is completely nonsensical to look at the difference between 500Hz and 8-9kHz, or 1.5khz and 8-9khz. Makes absolutely no sense.

I and others have had to correct you multiple times in this thread. Look, there's no nice way to say this, but if you're not sure what you're talking about, it's probably better to ask questions rather than pretend like you're "teaching" someone.

 

[Ageve]

OH WOW, you honestly do NOT know how to interpret off-axis response charts!!!!!! You really honestly don't know, yet you're pretending to analyze charts like you do.
Put simply, treble balance can be determined by the slope of the regression line through the FR curve. It should be plainly obvious that the slope is decreasing as you head more and more off-axis on the 805D4 charts you linked and referred to. Even people who disagree with my perspective on how B&W speakers sound can plainly see this in the charts.

And here you go again.

I didn't say a word about slope or regression line. You are trying your hardest to discredit me, but you are missing the point.

Again:

At 30 deg, the (804 D4) response is basically the same as on-axis, down to about 1.5 kHz. The tweeter level is reduced by 2-3 dB, but so is everything else above 1.5 kHz.

(Actually, it's the same all the way down to 300 Hz).

Please note: The thin green line is a copy of the 30 deg one. It's not rotated or manipulated in any way (other than by moving it). Look at the red vertical lines. The difference is basically the same, all the way from 300 Hz up to 9 kHz (different at the dip). Above 10 kHz, you can see reduced treble output (blue lines).

A flat response within 30 deg is obviously a good thing in itself, but it makes it difficult to tame the B&W tweeter by moving off-axis (to say the least). We're not talking about 1-2 dB differences here - It's way too bright.

In other words, 15-30 deg toe-in/out won't make much of a difference to the overall tonality (except for making the dip slightly better or worse depending on the angle).

I wasn't talking about overall slope and the fact that the tweeter output is reduced off-axis, at more extreme angles. It's obvious, and I didn't really think I had to mention it. The trend (dip followed by increased tweeter level) still remains though, even at 60 deg. It means that it will probably be possible to EQ them with good results, when/if Spinorama measurements are available. In fact, it would be very interesting to listen to them after this correction (both 804 D4 and the Batman speaker 805 D4).

If you want my honest subjective opinion (I'm sure you don't, but anyway), 802D is not a bad speaker at all, and it shouldn't be judged based on current models. I've listened to it a few times. It's a little on the bright side, and the dip is noticeable. The build quality is excellent. The pair matching was +/- 0.5 dB.

 

[preload]

And here you go again.

I didn't say a word about slope or regression line. You are trying your hardest to discredit me, but you are missing the point.

Sorry, I'm not letting you off the hook this time. This needs to stop.

You wrote: "It's not possible to tame the tweeter by moving off-axis (horizontal). Perhaps not surprising since the 805 D4 behaves in the same way." And then you linked to Soundstage measurements of the 805D4 (which are pasted below) as evidence. You then, for unclear reasons, attempted to compare FR chart levels at 500 Hz and 8-9 kHz, pointing out that the level difference doesn't change much off-axis (which is analysis that makes no sense).

You totally missed that the treble level dropped:
15 degs: -1dB over a 4 octave BW
30 degs: -3dB over a 4 octave BW
45 degs: -5dB over a 4 octave BW
But the thing is, I shouldn't even have to point this out because it should be patently obvious for someone to be able to successfully engage in this conversation.
I'm also bewildered by why you responded in a way that did not recognize why the slope of the FR regression line is so relevant here.

For everyone else who is capable of properly interpreting the off-axis response curves, the 805D4 charts shows how the brightness of that speaker clearly just be affected by the degree of toe-in.