New Bowers and Wilkins 800 series...

[徐中銳]

Official introduction videos :

 

 

 

[RFP]

I don't think that BnW where ever able to control the kevlar resonance at the top of the passband which to my ears sounds quite nasty. I owned the 802 S3 for a bit but the violins made my ears blead...

R

 

Surprisingly perhaps, the resonance is barely audible, from my experience anyway.

 

There's definitely no stridency on violins with the Diamond series, and if anything the upper mids sound a little too polite for my liking.

[BobP63]

I don't think that BnW where ever able to control the kevlar resonance at the top of the passband which to my ears sounds quite nasty. I owned the 802 S3 for a bit but the violins made my ears blead...

R

 

Rest assured this is certainly NOT the case with the 802 Diamonds. I'm curious to hear the new 802's but have no plans on parting with my diamonds anytime soon.

[RFP]

Rest assured this is certainly NOT the case with the 802 Diamonds. I'm curious to hear the new 802's but have no plans on parting with my diamonds anytime soon.

 

If the new 802D3's are at least the equal of my 800D2's, I might make the change but only because I would rather have a more compact speaker.

 

Fwiw, I'm inclined to think that the new D3 lineup represents change for the sake of change, more than real advancement, but I'll be happy to be proved wrong.

[Patrick Butler]

Hi RFP,

 

With the exception of the binding posts, the diamond dome and a few crossover components, the other 868 parts are completely new. Our engineers have been working on the project for over 7 years, and the advancements are very real.

 

Regards,

 

Patrick Butler

B&W Group North America

 

 

 

 

If the new 802D3's are at least the equal of my 800D2's, I might make the change but only because I would rather have a more compact speaker.

 

Fwiw, I'm inclined to think that the new D3 lineup represents change for the sake of change, more than real advancement, but I'll be happy to be proved wrong.

[cjf]

Hi RFP,

 

With the exception of the binding posts, the diamond dome and a few crossover components, the other 868 parts are completely new. Our engineers have been working on the project for over 7 years, and the advancements are very real.

 

Regards,

 

Patrick Butler

B&W Group North America

 

Patrick,

 

I would like to her your take on how the new 802 D3 and 803 D3 compare to each other if we ignore for just a minute the low frequency range differences between the two spekers?

 

What could a prospective buyer expect to hear between the two models in the frequencies not covered by the bass drivers? Would you say the differences in the non bass related categories would be equal to each other?

 

I ask this question because I am inclined to think one could easily make up for any bass diffiencies between the two models by adding a very good sub to the equation if one owned the lesser model so this only leaves the upper registers left to compare between the two models.

 

I am very seriously considering the new 802 D3 as a replacement for my current 803 D2 model but if the general consensus is that one could ultimatly end up with a better system by adding a sub to an 803 D3 setup some monies could possibly be saved in the process assuming the Mid range and Highs between the two models are equal.

 

Also, let's assume power to drive each speaker is a non issue and that X 4 MonoBlocs of equal power will be used in either scenerio.

 

Thanks for your time and insight on this matter.

[Patrick Butler]

Hi cjf,

 

Good question. Let's assume for the moment that you were to high-pass 803D3 and 802D3 at 80Hz. While I don't know the precise points where the woofers crossover to the mids, those points are probably around 300-350 Hz. Those two octaves above 80Hz are then covered by either the 802D3s larger woofers, or the 803D3s comparatively smaller ones. That makes a difference.

 

Then we get to the topic of the mids, where the 802D3 uses a 6" Continuum FST versus the 803D3s 5" Continuum FST. This also makes a difference.

 

Knowing all of this, when I listen to 802/803D3 back to back with male and female vocals, 802D3 sounds noticeably more present and palpable than 803D3. Even if you were to integrate a sub with 803D3, you still have that fundamental difference in upper frequencies (above 80Hz) where the sub is mostly not operating. The good news is that I would happily buy 803D3 over 802D2. The extra gains in clarity and reduction in overall levels of noise far outweigh the extra body you'll get in lower frequencies with 802D2.

 

Regards,

 

Patrick Butler

B&W Group North America

 

Patrick,

 

I would like to her your take on how the new 802 D3 and 803 D3 compare to each other if we ignore for just a minute the low frequency range differences between the two spekers?

 

What could a prospective buyer expect to hear between the two models in the frequencies not covered by the bass drivers? Would you say the differences in the non bass related categories would be equal to each other?

 

I ask this question because I am inclined to think one could easily make up for any bass diffiencies between the two models by adding a very good sub to the equation if one owned the lesser model so this only leaves the upper registers left to compare between the two models.

 

I am very seriously considering the new 802 D3 as a replacement for my current 803 D2 model but if the general consensus is that one could ultimatly end up with a better system by adding a sub to an 803 D3 setup some monies could possibly be saved in the process assuming the Mid range and Highs between the two models are equal.

 

Also, let's assume power to drive each speaker is a non issue and that X 4 MonoBlocs of equal power will be used in either scenerio.

 

Thanks for your time and insight on this matter.

[cjf]

Thank you Patrick for the detailed response.

[徐中銳]

Yesterday :

[Elberoth]

Wow, those speakers are ugly. I have never thought i would say sth like this about B&W speakers.

[徐中銳]

Wow, those speakers are ugly.

 

...

[RFP]

Hi RFP,

 

With the exception of the binding posts, the diamond dome and a few crossover components, the other 868 parts are completely new. Our engineers have been working on the project for over 7 years, and the advancements are very real.

 

Regards,

 

Patrick Butler

B&W Group North America

 

Thanks Patrick.

 

I very much look forward to auditioning the new range as soon as they hit the local dealer's showroom floor.

[徐中銳]

Some photos from the 2015 Tokyo International Audio Show :

 

 

 

[徐中銳]

News from Andrew Everard (and here if you'd like to read the Comments) :

Now *that's* a review sample...

 

 

And his complementary blog Post

[Sam Lord]

Hello Patrick,

 

I have one very specific and firm question about the new 800 series, and a couple of comments. I have found that few builders have the stomach to provide this information, and some discount it as unimportant. But I assure you that it matters!

 

1) I wish to know at what position, relative to the 802s, the listener's ears need to be for correct time alignment of the speaker's transducers. For a given degree of toe-in, there is ONLY ONE position possible for this to occur with speakers having more than two non-coaxial transducers having different frequency bands.

 

Now, if you *do* give a precise response, you get a gold star! Roger Quested ignored this question when put to him directly at an AES convention. The sad fact is that he didn't know!

 

Regarding ATC50s, Billy Woodman reluctantly admitted that they were designed for soffit-mounting, so converged *below* the axis of the woofer, and probably around 12-14ft. But the sad fact is, he didn't know either!

 

My former company only built loudspeakers with translating driver-enclosure adjustment, and would calculate for any owner the correct positioning of those drivers for time-alignment for any listening position. We concluded that group delays of >2usec in the treble (about 7 degrees at 10kHz) are audible, so precision *matters*. BTW Bruno Putzeys agrees fully with that figure and employs it in his active designs. One design, the large Focal, uses a ridiculous contraption that fails to telescope its drivers.

 

The ideal way to address the problem is to mount individual, adjustable driver enclosures within a frame, such as was done in the discontinued Intuitive Denali speakers which I helped design and calibrate. You can see errors in this prototype: protruding floors beneath each enclosure.

 

There is a subjective part to determining the correct distances to offset the different transducers...but in my experience when you set it right for your ears, the agreement on correctness is unanimous. Didn't early 801s have a movable tweeter for the purpose?

 

I believe that most builders are in error suggesting that heavy toe-in be used. When little or no toe-in is set, the first reflections have less group delay and act more like coherent echoes. And when time alignment of direct radiation is essentially perfect, even very quick first reflections and higher-order reflections vastly diminish in importance. Many of your mastering engineer clients agree. Ever try it yourself?

 

Finally, I want to congratulate your team for what appears to be a major advance in speaker design. The 802s with just a pair of 8" woofers are claimed to be flat to 17Hz, better than a half-octave improvement!!! I remember hearing a first-generation(?) pair of 802s driven by a Spectral preamp and a souped-up Conrad-Johnson tube amp, and heard *outstanding* fidelity...about 25 years ago.

 

Cheers, Sam

[Patrick Butler]

Hi Sam,

 

Good question. The reference axis for the new 800 Series is dead on, and slightly below the tweeter axis for optimal summing of the tweeter and mid. I always adjust rake to accommodate differing listening distances from the speakers, and listening height. I also find a bit of rake (back) helps to alleviate floor bounce cancellation at a given listening position. I have worked with some products that optimally sum at the tweeter's axis, all the way to designs that only work properly if you are laying on the floor.

 

Regarding toe-in I always use 10 degrees or more. This will depend on distance to side walls, distance apart and distance from the listener. In studios, I see everything from tweeters focusing directly on the listener at the console, to only 20 degrees of toe-in. Circling back to rake, assuming all other parameters of the setup are locked-in (a big assumption,) adjusting spikes for correct rake makes a huge difference. Coincident arrival is of paramount importance. Early 801 had an adjustable mid/tweeter module rotating in the horizontal axis (yaw), but the position of the tweeter relative to the mid was fixed.

 

Regards,

 

Patrick

B&W Group North America

 

Hello Patrick,

 

I have one very specific and firm question about the new 800 series, and a couple of comments. I have found that few builders have the stomach to provide this information, and some discount it as unimportant. But I assure you that it matters!

 

1) I wish to know at what position, relative to the 802s, the listener's ears need to be for correct time alignment of the speaker's transducers. For a given degree of toe-in, there is ONLY ONE position possible for this to occur with speakers having more than two non-coaxial transducers having different frequency bands.

 

Now, if you *do* give a precise response, you get a gold star! Roger Quested ignored this question when put to him directly at an AES convention. The sad fact is that he didn't know!

 

Regarding ATC50s, Billy Woodman reluctantly admitted that they were designed for soffit-mounting, so converged *below* the axis of the woofer, and probably around 12-14ft. But the sad fact is, he didn't know either!

 

My former company only built loudspeakers with translating driver-enclosure adjustment, and would calculate for any owner the correct positioning of those drivers for time-alignment for any listening position. We concluded that group delays of >2usec in the treble (about 7 degrees at 10kHz) are audible, so precision *matters*. BTW Bruno Putzeys agrees fully with that figure and employs it in his active designs. One design, the large Focal, uses a ridiculous contraption that fails to telescope its drivers.

 

The ideal way to address the problem is to mount individual, adjustable driver enclosures within a frame, such as was done in the discontinued Intuitive Denali speakers which I helped design and calibrate. You can see errors in this prototype: protruding floors beneath each enclosure.

 

There is a subjective part to determining the correct distances to offset the different transducers...but in my experience when you set it right for your ears, the agreement on correctness is unanimous. Didn't early 801s have a movable tweeter for the purpose?

 

I believe that most builders are in error suggesting that heavy toe-in be used. When little or no toe-in is set, the first reflections have less group delay and act more like coherent echoes. And when time alignment of direct radiation is essentially perfect, even very quick first reflections and higher-order reflections vastly diminish in importance. Many of your mastering engineer clients agree. Ever try it yourself?

 

Finally, I want to congratulate your team for what appears to be a major advance in speaker design. The 802s with just a pair of 8" woofers are claimed to be flat to 17Hz, better than a half-octave improvement!!! I remember hearing a first-generation(?) pair of 802s driven by a Spectral preamp and a souped-up Conrad-Johnson tube amp, and heard *outstanding* fidelity...about 25 years ago.

 

Cheers, Sam

[Sam Lord]

Hi Sam,

 

Good question. The reference axis for the new 800 Series is dead on, and slightly below the tweeter axis for optimal summing of the tweeter and mid. I always adjust rake to accommodate differing listening distances from the speakers, and listening height. I also find a bit of rake (back) helps to alleviate floor bounce cancellation at a given listening position. I have worked with some products that optimally sum at the tweeter's axis, all the way to designs that only work properly if you are laying on the floor.

 

Regarding toe-in I always use 10 degrees or more. This will depend on distance to side walls, distance apart and distance from the listener. In studios, I see everything from tweeters focusing directly on the listener at the console, to only 20 degrees of toe-in. Circling back to rake, assuming all other parameters of the setup are locked-in (a big assumption,) adjusting spikes for correct rake makes a huge difference. Coincident arrival is of paramount importance. Early 801 had an adjustable mid/tweeter module rotating in the horizontal axis (yaw), but the position of the tweeter relative to the mid was fixed.

 

Regards,

 

Patrick

B&W Group North America

 

Thank you Patrick, most illuminating. You provided the location of the horizontal plane, passing a little below the tweeter, for tweeter-midrange coherence, but what about distance? The woofer output needs to temporally match too. In a trial I made several years back with some 802D(1)s, I listened until I found a coherent zone at a sensible distance (somewhere around 12-14ft IIRC). Any of these calculations or experiments yields a nominal zone based on an ideal point: a space shaped like an ellipsoid that yields the most coherent direct radiation. So, can you provide a distance? Live sound speakers can converge at infinity but not monitors. My favorite test sample was a tom-tom hit on a drum record, maybe Sheffield's.

 

The reason I ask this to emphasize that fixed three-way speakers constrain the relative listening position in three dimensions. Most buyers don't think about how the geometry of three-way speakers limits the ideal listening distance as well as a narrow range of rake angle. It's not a criticism of 3-ways, but builders seldom mention it. Thank you again for your time and expertise.

[RFP]

Hello Patrick,

 

I have one very specific and firm question about the new 800 series, and a couple of comments. I have found that few builders have the stomach to provide this information, and some discount it as unimportant. But I assure you that it matters!

 

1) I wish to know at what position, relative to the 802s, the listener's ears need to be for correct time alignment of the speaker's transducers. For a given degree of toe-in, there is ONLY ONE position possible for this to occur with speakers having more than two non-coaxial transducers having different frequency bands.

 

 

Excellent point IMHO.

 

Fwiw, I've waved my test mic through the vertical lobe on-axis of my 800 diamonds and can verify that Patrick's statement above is correct for the D2's as it is the D3's, although the point of optimum flatness is about 10-15 degrees off axis with the D2 models.

 

However, to clarify, the point of optimum 'summation' is the point of phase coherence rather than 'time alignment'. Whether time alignment is audible or desirable is perhaps a point of conjecture, whereas phase coherence is most definitely a mandatory requirement.

 

Reading the previous D1/2 development paper, B&W's engineers had also deemed absolute polarity to be (more) important than maintaining a high degree of phase coherence > 0.5 octaves above and below the crossover frequency, as well as time alignment, and complicted by a preference for a simple crossover where the tweeter is high-passed using a simple first order electrical network.

 

The design imperatives chosen are not without some conflict and hence some compromise, so it will be intersting to see if the new D3 series crossover offers a different balance of compromise. In my view, the most negative consequence of the design decisions previously taken is the absolute necessity for the listener to be seated at exactly the right height and at a distance that's slightly greater than average for a large 3-way.

 

It's my view that many current D2 owners are not be getting the best from their

speakers because they're seated below the narrow lobe of pleasure.

[Axiom05]

It's my view that many current D2 owners are not be getting the best from their

speakers because they're seated below the narrow lobe of pleasure.

 

Being an owner of 802 D2 speakers (and N801's before that) I can speak from experience. The setup guidelines in the owner's manual are woefully inadequate. If, as you state, most people are seated too low (and I agree), then why did B&W place the tweeter/midrange so high? I had to purchase new chairs in order to get my ears a bit higher. And the problem is made worse when you add the spikes which raise the speaker up even more. With the spikes you can always tip the speaker forward to shift the vertical axis but you are working "blind" and, to be honest, the speakers look ridiculous tilted forward (not to mention scarily off-balance). I find the speakers sound best with a little toe-in (roughly half way between on-axis and zero toe-in) and with the listening position further away than the distance between the speakers. If you have specific set-up details that you would share, I am sure other owner's would be appreciative, particularly wrt the ear-tweeter/mid relationship. IIRC the 800 D2 tweeter is even higher!

[RFP]

Being an owner of 802 D2 speakers (and N801's before that) I can speak from experience. The setup guidelines in the owner's manual are woefully inadequate. If, as you state, most people are seated too low (and I agree), then why did B&W place the tweeter/midrange so high? I had to purchase new chairs in order to get my ears a bit higher. And the problem is made worse when you add the spikes which raise the speaker up even more. With the spikes you can always tip the speaker forward to shift the vertical axis but you are working "blind" and, to be honest, the speakers look ridiculous tilted forward (not to mention scarily off-balance). I find the speakers sound best with a little toe-in (roughly half way between on-axis and zero toe-in) and with the listening position further away than the distance between the speakers. If you have specific set-up details that you would share, I am sure other owner's would be appreciative, particularly wrt the ear-tweeter/mid relationship. IIRC the 800 D2 tweeter is even higher!

 

Yes, the 800D2 is even taller than the 802D2, and I had to modify my sofa legs to raise the height shortly after the purchase. When seated too low, a suck-out develops in the 1.5 - 2.5kHz region, which is confusing for many owners because their ears are close to the height of the FST-mid, which is where they're expecting maximum energy in that region.

 

I actually solved the problem with an engineering solution, so it's probably of little use to most owners because it involves a small modification to the crossover itself, which is beyond most people's inclination, let alone capability.

 

Re-engineering the crossover isn't something to be taken lightly of course, and it involved many hours of measurements including individual driver measurements, impedance sweeps, and endless hours of computer simulations etc.

 

The end result is that I've added a quite simple 3 component R-L-C network in parallel with the tweeter which tilts the beam down very slightly but the main purpose is to align the phase of the tweeter and FST mid in the 1-3 KHz region, so that the sweet spot is widened out considerably, to the point where listening height is far less critical below the tweeter height. The 2kHz dip is thus all but eliminated. Subjectively, the speaker is more 'alive' in the mids and there's more snap, focus, and projection into the room. In layman's terms, the drivers are working more with each other instead of against each other.

 

As a past owner of the N801, you would be aware that the N800 series didn't suffer from the problem to the same extent. The main reason for that is that the N800 series used a 3rd order electrical filter on the tweeter instead of a first order filter, so there was far less driver overlap, and the drivers were actually more time aligned too which improves phase coherence above and below the M-T crossover frequency, thus making the main lobe wider.

 

My modified tweeter filter response brings it closer to 2nd order electrical than first order electrical, although interestingly the measured response of the tweeter now better represents the target 2nd order L-R. Another benefit is that the tweeter coil resonance is effectively damped and power dissipation is reduced.

[RFP]

Measured at the sofa height and about 1.8m from the speaker, with and without the crossover mod.

[Axiom05]

Measured at the sofa height and about 1.8m from the speaker, with and without the crossover mod.

 

Yes, it looks good.

[徐中銳]

Starting with his €160,000 purpose-built listening room...

[Patrick Butler]

Hi Sam,

 

Apologies for the late reply. Been on the road this week with long days and late nights.

 

The minimum listening distance for 802D2 or 802D3 would be 9' 10". That is where all the drivers align. You can listen further away if wished (and some definitely do,) but considerably closer will result in the drivers not aligning to their full potential.

 

 

Regards,

 

Patrick

B&W Group North America

 

 

 

Thank you Patrick, most illuminating. You provided the location of the horizontal plane, passing a little below the tweeter, for tweeter-midrange coherence, but what about distance? The woofer output needs to temporally match too. In a trial I made several years back with some 802D(1)s, I listened until I found a coherent zone at a sensible distance (somewhere around 12-14ft IIRC). Any of these calculations or experiments yields a nominal zone based on an ideal point: a space shaped like an ellipsoid that yields the most coherent direct radiation. So, can you provide a distance? Live sound speakers can converge at infinity but not monitors. My favorite test sample was a tom-tom hit on a drum record, maybe Sheffield's.

 

The reason I ask this to emphasize that fixed three-way speakers constrain the relative listening position in three dimensions. Most buyers don't think about how the geometry of three-way speakers limits the ideal listening distance as well as a narrow range of rake angle. It's not a criticism of 3-ways, but builders seldom mention it. Thank you again for your time and expertise.

[Axiom05]

The minimum listening distance for 802D2 or 802D3 would be 9' 10". That is where all the drivers align. You can listen further away if wished (and some definitely do,) but considerably closer will result in the drivers not aligning to their full potential.

 

Well, at least I got the listening distance right!

 

Patrick, do you have any tips on exactly how to adjust the speaker in the vertical plane? How do you choose a specific amount of downward/forward tilt to compensate for lower than optimum listening height? I purchased higher chairs but I am afraid that the increase in chair height was negated by the increase in speaker height when I put the feet on. I couldn't get the speakers level on my tile floor with the castors as one or two castors were always sitting in the groove between adjacent tiles, therefore I had to resort to using the feet.

 

Cheers,

Bob