• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

B&W 800D4 series

preload

Major Contributor
Forum Donor
Joined
May 19, 2020
Messages
1,554
Likes
1,701
Location
California
Its not rocket science, we understand its behaviour and the resulting voicing which some people like and some don't.

There's a difference between being able to "characterize" the behavior via measurements and "understanding" how those measurements translate into perceived sound quality. Many individuals seem to carry a rigid belief that only measurements that match the Harman model can produce high quality sound. It still hasn't been reconciled how, as you point out, many people prefer the B&W 800 sound signature, yet it breaks all of the Harman rules when it is measured.
 

preload

Major Contributor
Forum Donor
Joined
May 19, 2020
Messages
1,554
Likes
1,701
Location
California
That on-axis treble gets reflected in 90% of rooms and makes the whole listening experience ’bright’ whether you’re on or off-axis from the speaker …few of us have massive ugly soffits and GIK panels in their Livingroom

it would be nice if Bowers had another set of terminals to the tweeter crossover which gave you the option of a few dB less (for arguably bright rooms) if you plugged your jumpers into those… so I wish… ;-)

I've had a pair of 802D speakers in 2 different living rooms, and they never registered as "bright." Like many loudspeakers, you can change the degree of toe-in to alter what is direct vs reflected. I hope you understand that reflected sound has a lower SPL than direct sound, and in some cases (when it is distant), the reflected sound doesn't even register as part of the original sound.
 

lherrm

Active Member
Joined
Nov 1, 2020
Messages
203
Likes
378
Location
Reunion Island
They messed up the volume matching, unfortunately. The Grimm LS1be was played ~4dB higher than the D&D 8c, which no doubt contributed to it sounding so much better. Not that it wouldn’t have still won(I really don’t know), but I do think the volume mismatch is responsible for most of the difference.

I believe they corrected this and posted the volume matched audio somewhere on their site. Unfortunately, I haven’t been able to find that audio, and they never responded to my question of where to find it :(.
.

There : https://alpha-audio.net/review/trip...monitoren-met-samples-en-video/?highlight=LS1

The samples are supposed to level match :


1631728609121.png


Not so sure when I listen to them.
But yeah, I like the LS1's better.
 
Last edited:

thewas

Master Contributor
Forum Donor
Joined
Jan 15, 2020
Messages
6,751
Likes
16,188
There's a difference between being able to "characterize" the behavior via measurements and "understanding" how those measurements translate into perceived sound quality. Many individuals seem to carry a rigid belief that only measurements that match the Harman model can produce high quality sound. It still hasn't been reconciled how, as you point out, many people prefer the B&W 800 sound signature, yet it breaks all of the Harman rules when it is measured.
In the end no model will probably be ever be able to fully predict individual human subjective preference, one example is habituation, somebody who is listening for years with extra treble (not referring to B&W here) will likely find a neutral signature strange and dull for the fist period. What a correlation model like the Harman one can mostly do is a give a statistic probability of the preference in a population group.
 

Kvalsvoll

Addicted to Fun and Learning
Audio Company
Joined
Apr 25, 2019
Messages
878
Likes
1,643
Location
Norway
They messed up the volume matching, unfortunately. The Grimm LS1be was played ~4dB higher than the D&D 8c, which no doubt contributed to it sounding so much better. Not that it wouldn’t have still won(I really don’t know), but I do think the volume mismatch is responsible for most of the difference.

I believe they corrected this and posted the volume matched audio somewhere on their site. Unfortunately, I haven’t been able to find that audio, and they never responded to my question of where to find it :(.

I’m actually very interested in making recordings like this for my own personal use, to help hear relative difference between some of the speakers I own. I know absolute quality can’t be judged via recording, but I’m only concerned with relative differences in tonality (ex: A has more 1-4kHz energy than B). JBL 708p vs 308p is the comparison I’m most interested in atm, as surprised at how tonally different the sound with near identical waveguide, size, shape, and with similar measurements. What kind of mic do you think would work best for this purpose? Right now, I just a Umik(1 and 2), but I don’t mind spending a bit of money if it would help.

Just had another quick listening to this, and I have to say, it works quite well - differences are possible to hear, and none of the recordings sound horrible compared to original.

A good mic for this purpose needs to have - obviously - a flat and full frequency response, and the right directivity characteristics. A measurement mic like the umik does not work well, because it is omni, you will only get the bass right, up in frequency there will be too much room contribution. And then there is the mic placement - x-y or separated, listening position or closer. Some have chosen binaural, not so sure that is the best. But here we have an example of something that seems to work quite well, so why not just copy what they did here - get the same mics, place them like they did.

So how do these recordings compare to the original? Well, there is better spatial contrast, generally better resolution and definition in the high freqs, bass is quite different, and all of them lost this immediate impact on the drums that you have in live instruments. Differences in the bass is expected - no system without a calibrated, separate bass-system can compete here. Overall, I found the d&d to be most true to the original, even though they had a lift down in the bass range.
 

Art Vandelay

Member
Joined
Apr 29, 2020
Messages
29
Likes
32
Detailed measurements of the 800D3 are available (all images are from there). The D4 should not behave much differently in comparison.

Excellent values for harmonic distortion and intermodulation distortion. The speaker's decay behavior is also impeccable. The maximum achievable sound pressure is excellent. There is not much to write about these points.


Contrary to what is discussed here, at least the 800D3 clearly shows the behavior of a loudspeaker that is crossed at both crossover frequencies with fourth-order filters.
Due to the classic BR tuning around 270° phase rotation in the low frequency range (blue marked) and at the crossover frequencies at 500Hz and 3.5kHz each further 360° phase rotation (red and green marked), as one would expect for fourth-order filters.


Those who consider a low group delay important will not be satisfied by this speaker. Due to the crossover frequency at 500Hz and fourth-order filter, the group delay in the midrange will be unusually high.

That's interesting, because the D2 model is definitely 2nd order M-T.

For what it's worth, the much revered Salon2 also employs 4th order filters and with crossover points at 150, 575, and 2300, so group delay will look even worse due to the extra crossover point.
 

ctrl

Major Contributor
Forum Donor
Joined
Jan 24, 2020
Messages
1,616
Likes
6,086
Location
.de, DE, DEU
For what it's worth, the much revered Salon2 also employs 4th order filters and with crossover points at 150, 575, and 2300, so group delay will look even worse due to the extra crossover point.
Yep, fourth order filter are standard and you are right about the group delay. That's why I wrote "those who consider GD important...".

Even using these 3-way crossovers, the audibility threshold for GD with artificial signals is just about reached or slightly exceeded - according to current research AFAIK.
How important this is, everyone may decide for themselves.
 

MaxBuck

Major Contributor
Forum Donor
Joined
May 22, 2021
Messages
1,515
Likes
2,118
Location
SoCal, Baby!
If you like these speakers and own them, I'm happy for you. With that said, I'm glad I don't like their sound, because I think they look absolutely hideous.
 

atsmusic

Active Member
Joined
Sep 10, 2021
Messages
109
Likes
85
If you like these speakers and own them, I'm happy for you. With that said, I'm glad I don't like their sound, because I think they look absolutely hideous.

what model of the d4's did you listen to?
 

richard12511

Major Contributor
Forum Donor
Joined
Jan 23, 2020
Messages
4,335
Likes
6,700
Detailed measurements of the 800D3 are available (all images are from there). The D4 should not behave much differently in comparison.

Excellent values for harmonic distortion and intermodulation distortion. The speaker's decay behavior is also impeccable. The maximum achievable sound pressure is excellent. There is not much to write about these points.

The crossover frequency of the woofer and midrange driver is unusually high at 500Hz and the distance from the lower woofer to the midrange driver is quite large, which creates a constriction in the radiation in the vertical direction.
The crossover frequency from midrange to tweeter is very high at 3.5 kHz, but works better than expected in terms of horizontal directivity.


The horizontal directivity taken as a whole is rather poor. I have drawn the course of the -6dB sound pressure limit (deviation in relation to the axis frequency response) here:
View attachment 153526
The 7kHz widening is caused by a dip in the axis frequency response and should not be a problem.

In the frequency range above 1.2kHz, the horizontal radiation is +-45°, which could almost be called narrow.

Quite bad is the abrupt change of the radiation at frequencies below 1.2kHz. Within one octave, the horizontal dispersion changes from +-35° (around 2.5kHz) to +-110° (around 1.2kHz) - but to be completely sure, one would have to look at the individual frequency responses.

The vertical listening area is very limited due to the high crossover frequency. Upwards, clearly less than 10° until the sound pressure deviates more than 6dB from the axis frequency response. Downwards it should be around 10°.
View attachment 153529
It would certainly be very interesting to have a look at the power response and PIR of this speaker.


Contrary to what is discussed here, at least the 800D3 clearly shows the behavior of a loudspeaker that is crossed at both crossover frequencies with fourth-order filters.
Due to the classic BR tuning around 270° phase rotation in the low frequency range (blue marked) and at the crossover frequencies at 500Hz and 3.5kHz each further 360° phase rotation (red and green marked), as one would expect for fourth-order filters.
View attachment 153531

Those who consider a low group delay important will not be satisfied by this speaker. Due to the crossover frequency at 500Hz and fourth-order filter, the group delay in the midrange will be unusually high.

The directivity errors are expected given the design choices, but I'm surprised at how narrow the overall dispersion is, especially with their essentially open baffle tweeter/mid design.

The cabinet decay behavior is perhaps the best I've ever seen, and the new model should be even better in that regard :D, due to the aluminum treble/mid enclosures.
 

atsmusic

Active Member
Joined
Sep 10, 2021
Messages
109
Likes
85
Not the D4s; I've listened to the D3s and many previous B&Ws. They have a consistent "house sound" that I don't care for. Obviously many people love it.

I hear that a lot on here. What is the house sound they have? And all the nautilus series have it?
 

Vear

Member
Joined
Dec 28, 2019
Messages
33
Likes
50
I have 805D2's and I love them, they sound like reality to me.
My hearing is fine and yes I've had other speakers (stand-mounts) that supposedly measure flat yet they often sound compressed and thick, not what real life sounds like to me.
 
Last edited:

Art Vandelay

Member
Joined
Apr 29, 2020
Messages
29
Likes
32
Yep, fourth order filter are standard and you are right about the group delay. That's why I wrote "those who consider GD important...".

Even using these 3-way crossovers, the audibility threshold for GD with artificial signals is just about reached or slightly exceeded - according to current research AFAIK.
How important this is, everyone may decide for themselves.

Just to clear up one point. The tweeters on 800D1-3 series are all a physical half wavelength advanced from the FST mid (at crossover F), so that being the case you will expect to see a 360 degree phase rotation - with 2nd order filters. This also explains the slight suckout on axis at 7-8 kHz, because at 2x crossover frequency the tweeter is advanced a full wavelength and is hence anti-phase with the FST. Because the FST is highly directional at7kHz the suckout only appears on axis +/- 10 degrees.
 

ctrl

Major Contributor
Forum Donor
Joined
Jan 24, 2020
Messages
1,616
Likes
6,086
Location
.de, DE, DEU
Just to clear up one point. The tweeters on 800D1-3 series are all a physical half wavelength advanced from the FST mid (at crossover F), so that being the case you will expect to see a 360 degree phase rotation - with 2nd order filters. This also explains the slight suckout on axis at 7-8 kHz, because at 2x crossover frequency the tweeter is advanced a full wavelength and is hence anti-phase with the FST. Because the FST is highly directional at7kHz the suckout only appears on axis +/- 10 degrees.

You quote from the white paper of the 800 series, which I had completely overlooked.

This explains the phase frequency response with 360° phase shift around the crossover frequency (measured in the review) at around 4kHz although only second order filters slopes are used. A clever approach to avoid the polarity reversal of a driver, when using second order filters, at the expense of the axis frequency response (and phase shift).
Of which B&W is convinced, because they write:
Reverse polarity connection of drive units has always been avoided at B&W in recent years, it being felt that the sound lacked coherence and focus.


To get a feel for what the power frequency response and PIR might look like for such a crossover, I adapted the Directiva DIY speaker accordingly.
The single drivers have a fairly even sound pressure radiation, which is why the result looks good above and far below the crossover frequency.

As with the 800D3, the tweeter and woofer are offset by half a wavelength of the crossover frequency.
The similarities in frequency response (top left image) are clearly visible. The vertical normalized sonogram (bottom right) also looks almost identical to the measurement in the review.
1631798569845.png
For the sound power (center left image, blue curve) and PIR (orange curve), however, there is a pronounced dip in the 2-4 kHz range - this should also look similar with the 800D3.
This and the relatively narrow directivity should account for the B&W "house sound".
 

Art Vandelay

Member
Joined
Apr 29, 2020
Messages
29
Likes
32
For the sound power (center left image, blue curve) and PIR (orange curve), however, there is a pronounced dip in the 2-4 kHz range - this should also look similar with the 800D3.
This and the relatively narrow directivity should account for the B&W "house sound".

Precisely.

It's entirely a consequence of the design approach taken and yes, will inevitably result in a 'house sound'.

The dip in the 2-3 kHz area also mirrors what some call the 'BBC' or 'Gundry' dip, which is discussed elsewhere in this forum.

Noting that the dip increases away from the center axis and is most pronounced at +/- 90 degrees, where the FST and tweeter are aligned in time and hence where the phase error between M and T is greatest, and also below tweeter axis for the same reason.

When most people look at the polar response they (mis)interpret the result as a consequence of a 'too large diameter' mid driver exhibiting excessive beaming.


On axis - 802D3...
1631802887899.png
 
Last edited:

napilopez

Major Contributor
Forum Donor
Joined
Oct 17, 2018
Messages
2,109
Likes
8,424
Location
NYC
While we're at at it, I think it's worth remembering/highlighting that old B&W speakers, even those that also had floating tweeter designs, exhibited much cleaner on-axis and directivity, so it's really not fully a limitation dictated by aesthetics. The response in modern B&Ws is absolutely a purposeful choice. Although I disagree with some of preload's thoughts earlier, it should indeed be obvious that B&W could make a 'harmaneseque' speaker if they wanted to. They already did so decades ago.

Here's the B&W Matrix 801 Series 2, as measured by stereophile way back in 1987:

1631803923590.png


1631803943738.png


1631803965017.png


This 34 year old speaker measures better by my standards than anything recent from the company. Not saying it'll necessarily sound better than their newer stuff, but again, just reinforcing the obvious that they could make speakers with better directivity and more linear on-axis if they wanted to.

Another example, the CDM1, reviewed in 1996:
1631804971579.png


1631804992717.png


Same story.

The original B&W 705, measured in 2002:

1631805179238.png
1631805147971.png


The B&W Nautilus 805, measured in 1999:
1631805502362.png


1631805587265.png

B&W Matrix 805, measured in 1993:
1631806147361.png

1631805983432.png


1631806042270.png


It's also worth noting that B&W seems to have largely built its modern reputation and mindshare on speakers with this kind of performance, not on its current more esoteric speakers.

On the discussion about their partnership with Abbey Road, that partnership began way back in 1988, when B&W speakers measured like this. Whatever way you look at it, it was a completely different sound and tonal balance. So either they were more neutral back then, or theyre more neutral now (again, not even talking about what's 'better,' just basic timbral neutrality). You can't have it both ways.

(Personally, I'm quite sure the current partnership is more about marketing than anything else. B&W is arguably the most prestigious speaker company in the world in terms of wider mindshare among audiophiles.)

My impression of modern B&W is that they maximize technical performance in driver design and many things outside of FR and directivity. I'm personally fully convinced that their current tuning tendencies have more to do with 'standing out' than optimizing sound for a majority of listeners, and that the company's engineers have likely focused on specific engineering goals (lowering distortion, power handling, etc) that don't fully prioritize tonality. Forest and trees and all that.

That's fine, mind you, as long as listeners know what they're getting.

Edit: This is reinforced by the fact that there doesn't seem to be any real consistency in how modern B&W speakers measure (Or in my personal experience, which is just with the modern stuff, how they sound).
 
Last edited:

MarcT

Addicted to Fun and Learning
Forum Donor
Joined
Nov 14, 2019
Messages
931
Likes
612
Location
East Texas
Precisely.

It's entirely a consequence of the design approach taken and yes, will inevitably result in a 'house sound'.

The dip in the 2-3 kHz area also mirrors what some call the 'BBC' or 'Gundry' dip, which is discussed elsewhere in this forum.

Noting that the dip increases away from the center axis and is most pronounced at +/- 90 degrees, where the FST and tweeter are aligned in time and hence where the phase error between M and T is greatest, and also below tweeter axis for the same reason.

When most people look at the polar response they (mis)interpret the result as a consequence of a 'too large diameter' mid driver exhibiting excessive beaming.


On axis - 802D3...View attachment 153682
Lol, it looks to me like the rise at 10k is two or three times the dip at 2-3k! Which of those deviations is more likely to be noticeable, or problematic, for the average audio enthusiast in a mostly untreated domestic listening environment? Just asking because I'm starting to kick around the idea of picking up a set of B&W because of their dynamic capabilities and very low distortion.

BTW, there are a number of used 804 D3 available at really good prices right now. Is that because people decided they don't like them, or do you think people are unloading them to get into the D4 series?
 
Last edited:
Top Bottom