Going Dutch (& Dutch): a Conversation with Martijn Mensink Page 2

Atkinson: The Stereophile review's measurements showed that the 8c has a time-coincident step response, which again is extremely unusual. While there are passive dynamic speakers that achieve this on the listening axis with sloped-back front baffles and first-order crossover filters, there are very often compromises as you move away from that axis.

Mensink: Yes. You need a DSP to do that properly . . . As the tweeter and the midrange driver are on the front baffle, the woofers on the back, this introduces a problem. Normally the tweeter and midrange signals would arrive at your ears before the bass does. The tweeter and midrange themselves are acoustically aligned because they are in the same plane, but then we delay them a little to virtually align them with the woofers in the back. It becomes a virtual point source.

But when you place the speakers in front of a wall, as the tweeter and the midrange driver are directional, acoustically speaking it's as if the wall really isn't there. However, as the woofers are on the back, the wall is sort of an acoustic mirror, so it's as if there's another pair of woofers behind the wall. In effect you now have four woofers, and that means that their combined acoustic point of origin shifts in the direction of the wall. We add a bit of extra delay to the tweeter and midrange driver to compensate for that, so that in combination with the wall, again you have a virtual point source.

Atkinson: And you have a reasonably wide range of distances you can compensate for, up to 1.5m/5'?

Mensink: Yes, with the app you can tell the speaker "Alright you're 40cm from the front wall, you're 1m from the sidewall," then the speaker will automatically load the corresponding DSP presets.

420dutch.2

Atkinson: Then, as I describe in my follow-up review, using REW you can bring the acoustic response in the room as close as possible to the target curve that you've defined. Now you have a speaker where, not only is its intrinsic behavior optimized as a point source, its interaction with the room is also optimized.

Mensink: Yes, exactly, we think that is the final frontier . . . what we and a lot of other companies have done and are doing, we're taking Floyd Toole's lessons and incorporating them in designing a better loudspeaker. However, the room-loudspeaker interface is the final frontier and while there are other companies who implement DSP to deal with the room, the 8c does more than that in that it also is a cardioid, so it interferes with the room in a different way, a favorable way. I think that is the future.

Atkinson: How would you make a better 8c?

Mensink: I'd say the only compromise is loudness. The 8c is actually loud enough for most people but it's a relatively compact speaker. It's pretty loud but there are limits, like every speaker has its limits . . .

Atkinson: Like with the excursion you need from the woofers to get extension down to 20Hz?

Mensink: Yes, of course, [which is why] it would help a lot if you place the speaker closer to the front wall because it would augment the bass . . . There are different approaches that could work very well as well but I think it would be difficult to improve the 8c as it stands as a concept by a whole lot. Perhaps there's a small thing here, a small thing there that we could do but there's not a whole lot we could do, I think.

420dutch.anachoic

Atkinson: One thing you have been working on the 8c's integration with Roon when the pair of speakers are connected to a network.

Mensink: It took us a long time to get these speakers Roon Ready. And that's not because of Roon, because Roon is very easy to implement, it took our guys less than a week to make the 8c pretty much Roon Ready, to have the RAAT [(Roon Advanced Audio Transport)] protocol implemented. But the difficulty is what Roon does: it streams a digital stream to a streamer and there [the stream] is split into Left and Right channels.

Atkinson: But the 8cs are two separate units. Roon is not sending one stream going to both left and right speakers, it is one stream going to the left, the other going to the right, as if they were two separate multi-room zones.

Mensink: Correct. And each speaker has its own audio clock, which are not exactly the same. There's a little bit of drift but you don't notice that when you're not in the same room. And while Roon can send a stream to one speaker and a stream to the other speaker and you press "Play," they start more or less at the same time and it sounds okay playing music, a central voice would start to drift . . .

Atkinson: As the clocks very slowly lost sync with one another . . .

Mensink: Correct. So after an hour of playing the voice would be over there, not in the middle. It actually took a lot of time and a lot of work to get a pair of 8cs perfectly sync'd over the Ethernet. So that was the big hold-up.

Atkinson: That problem has been solved.

Mensink: Yes. The guys at Roon are currently testing the latest build, but it's not released yet. We're close though. And now that problem is solved, that enables us to add other features, other streaming services as well. We're big believers in Roon but there are other streaming services and so that is the future we are looking at.

Atkinson: Future plans?

Mensink: Well, it is our policy not to disclose anything but I can say a couple of things. A new version of the app, a native app that you can use on your phone, for instance.

The future we see is basically you will have just a pair of speakers and nothing else. You won't need any other gear. And we believe that everything will be streaming so you will no longer have a DAC or a streamer box, something like that. Just a pair of speakers and an interface. And everything will be streaming . . .

Atkinson: I think it is the future: smart speakers and a dumb source of data.

Mensink: Yes, exactly. In the old days, the glory days of hi-fi, the R&D and the innovation would be at the high end and then it would gradually trickle down to the lower-price gear, But now most of the innovation is in the small Bluetooth speakers—very cheap, sound actually quite decent. If you look at what it is it's remarkable and convenient and . . . well hi-fi's still pretty traditional, most guys still have a CD player and now they are gradually making the transition to streaming, that's a good thing . . .

Atkinson: A final thought?

Mensink: We don't think there needs to be this schism between audiophiles and normal people . . . we hope to bring them together.

ARTICLE CONTENTS

COMMENTS
Bogolu Haranath's picture

Audiophiles have ears like 'Yoda' ....... Normal people don't have those :-) ........

Bogolu Haranath's picture

D&D could consider making floor-standing speakers like the Eikon Audio Image1 integrated system :-) ......

Kal Rubinson's picture

Alternatively, they could make a woofer enclosure in the form of a stand that would integrate esthetically and acoustically with the 8c. :-)

Bogolu Haranath's picture

Like Kii Three + BXT ...... In that case, the woofers in the back of the cabinets of 8c would be superfluous, but they could also be used to augment the bass ....... They can also be sold separately and the previous owners of 8c could upgrade, if they wanted to ...... You (KR) could talk to those guys at D&D :-) .......

Bogolu Haranath's picture

The price of 8c + powered woofer stands should be competitive with the $25k price of Eikon Image1 system :-) ..........

Kal Rubinson's picture

...... You (KR) could talk to those guys at D&D :-) .......

I spent the better part of a day and evening talking with Martin during his visit to NYC. We talked about this and a lot of other stuff.

Bogolu Haranath's picture

Great ..... Waiting for a follow-up review ....... May be even another D&D new product review :-) .......

blang11's picture

I'm very much on board with what Dutch & Dutch is doing. Cheers to Stereophile for the continued coverage of the 8C.

Michael Fremer's picture

You realize that when you hear whiners say "the reason high end audio is drying up...." and "it's all a bunch of old men...." it's total nonsense.....

Anton's picture

I hadn't thought of it 'til you mentioned it.

Darned tootin' right!

From that pic, we can see it's only "half old men."

The 'whiners' got the "men" part dead right, though.

;-D

They share the same gender diversity as the Senate Committee for Women's Health!
https://www.google.com/url?sa=i&url=https%3A%2F%2Faction.momsrising.org%2Fsign%2Fahca_womens_health%2F&psig=AOvVaw1t145_l2YO5YPYa7LbTkA1&ust=1586646052969000&source=images&cd=vfe&ved=0CAIQjRxqFwoTCODGrKf73ugCFQAAAAAdAAAAABAN

Not being partisan, just trying to cherchez la femme!

Bogolu Haranath's picture

They look like members of House 'Diversity and inclusion committee' :-) ........

Bogolu Haranath's picture

They look like they just received 'immunity certificates' :-) ..........

Ortofan's picture

... are:

1) Do the amplifiers built into the 8c perform as well (or better) than the Benchmark AHB2?

2) Why was aluminum chosen for the LF and MF driver cones?
Research performed by Harbeth supposedly determined that the plastic formula they developed has the most ideal properties for a loudspeaker driver cone.

Bogolu Haranath's picture

Manufacturers like Revel, KEF and some others use aluminum, aluminum alloy and titanium metal cones ...... They measure well and also sound great :-) .......

Bogolu Haranath's picture

Manufacturers like Revel, KEF and some others use aluminum, aluminum alloy and titanium metal cones ...... They measure well and also sound great :-) .......

Bogolu Haranath's picture

In most of the 'active' speakers, the amplifiers are directly connected to the drivers ....... There is no crossover in between ....... So, the distortion levels of both the drivers and the amplifiers, should be very low ...... There is no 'middle-man' crossover, interfering and making up the rules, so to speak ......... Also, the crossover itself is usually active ...... Active crossovers usually have low distortion compared to passive crossovers ...... DSP control further enhances the entire setup .........

Active DSP controlled speakers are the future, and are gonna dominate the market :-) .......

Bogolu Haranath's picture

It is amazing that, $12,500 D&D speakers can match the performance of over $60k Magico M2 ....... See, JA1's listening room measurements ........ That's what active DSP controlled speakers can do :-) ...........

Bogolu Haranath's picture

The total cost of M2 + Lamm amps + speaker cables is close to $100k :-) ........

John Atkinson's picture
Ortofan wrote:
1) Do the amplifiers built into the 8c perform as well (or better) than the Benchmark AHB2?

There isn't any way of knowing, as you can't access the amplifiers' inputs and outputs.

Ortofan wrote:
2) Why was aluminum chosen for the LF and MF driver cones? Research performed by Harbeth supposedly determined that the plastic formula they developed has the most ideal properties for a loudspeaker driver cone.

To operate as a pure piston within its intended passband, a drive-unit cone should be very stiff, as well as being lightweight. Aluminum's high stiffness pushes the cone's breakup modes sufficiently high in frequency that they will be eliminated by the crossover's low-pass filter. (Easier to write than do.)

Non-metallic cones, like paper and plastic go into breakup mode lower in frequency, and no longer behave pistonically over the entire passband required. So, to prevent the breakup modes from coloring the sound, the material needs to have high internal damping. (Again, easier to write than achieve.) This is what I understand Harbeth's statements to mean.

John Atkinson
Technical Editor, Stereophile

Ortofan's picture

... I would have liked to have heard what Martijn Mensink had to say in regard to speaker cone materials. Also, he should have been able to address the performance of the amplifiers they use.

KEF's designer, for example, opts for aluminum cones and believes that with appropriate selection of crossover frequency and slope that any cone breakup and resonance will not be a factor, whereas Harbeth's designer apparently believes that some trade-off between ultimate cone stiffness and damping is worthwhile.

I'd also like to hear Andrew Jones thoughts on the subject, given that ELAC uses different types of cone materials in their product line, including aluminum and aramid fiber. Also, Pioneer, with whom Andrew Jones was previously associated, makes some speakers with injection molded plastic cones, as does Harbeth.

Bogolu Haranath's picture

Aha ...... We found another use for recycled plastic ...... Just joking :-) ........

Bogolu Haranath's picture

Several of Class-D amps measured by Stereophile have SNR better than 110 db (19 to 20 Bits) ....... The new Purifi Class-D amp specs show close to 120 db SNR (20 Bits) ....... The amplifier performance also depends upon how much negative feedback the designer chose to use ........ Amplifier performance also depends upon what impedance the speaker driver(s) have .........

In case of active speakers, the specific amplifier performance can be optimized for the specific driver, because they are connected directly :-) .........

Ortofan's picture

1) the manner in which amplifier performance depends upon how much negative feedback the designer chose to use.

2) the manner in which amplifier performance also depends upon the impedance of the speaker driver(s).

3) the manner in which amplifier performance, in case of active speakers, can be optimized for the specific driver because they are connected directly.

Bogolu Haranath's picture

A general purpose amplifier designer chooses the negative feedback, which he/she thinks works best with many partnering loudspeakers and listener preferences ....... The distortion levels depend accordingly ........ If you look at the measurements of general purpose amplifiers, Class-D or otherwise, you can see the various distortion levels .......
General purpose amplifiers designed for the usual 8 or 4 Ohm loads, distort if 2 Ohm loads are presented, or they just shut-down .......
In case of active speakers, the individual drivers can be designed for ideal impedance and their individual partnering amplifier can be designed accordingly with proper negative feedback and impedance load :-) .........

Ortofan's picture

... some correlation exists between the value of speaker impedance and the appropriate amount of feedback to be designed into the driving amplifier?

If so, do provide further details.

Also, given that the impedance of most speakers is not a constant value, but varies over the operating frequency range, are you therefore implying that the amount of feedback designed into the driving amplifier should also not be constant across its operating frequency range?

Bogolu Haranath's picture

The usual dynamic drivers used in the active speakers present the partnering amplifier with variable impedance as the voice coil changes position in the magnetic gap ....... Since each driver is connected to one or more amplifiers, they can be considered as a single unit and can be designed more efficiently, with appropriate feedback control of the amplifiers for the connected driver and desired sound quality ........ There is no interference from the crossover network and interference from other drivers operating at different frequencies ....... Active speaker design is a better and more efficient way of doing things :-) .....

Ortofan's picture

... "appropriate" feedback control of the amplifiers for the connected driver and desired sound quality?

Bogolu Haranath's picture

No rules ...... If you want the speaker driver(s) to sound like they are being driven by tubes, use low negative feedback, meaning high output impedance of the partnering amplifier(s) :-) .......

Ortofan's picture

... having some relatively low level of negative feedback (and, thus a relatively higher output impedance) will ensure that a solid-state amp sounds like a tube amp?

If so, exactly what level of negative feedback is required to achieve that effect?

Will that effect still be audible if the speakers employ conjugate load matching?

Bogolu Haranath's picture

No .... I'm not implying that a SS amp can or will sound like a tube amp, with low negative feedback ...... It may sound less 'transistor-y' with less negative feedback ........ I don't know how that affects speakers which employ conjugate load matching ....... Do you have any examples of such speakers reviewed by Stereophile with measurements? :-) ........

Bogolu Haranath's picture

By 'cojugate load matching', do you mean, dynamic speakers with pure resistive load and no phase angles? :-) .......

Bogolu Haranath's picture

See, JA1's review of KEF Ref.207/2 speakers ........ That 207/2 model seems to be using 'conjugate load matching' in the bass frequencies ........ That speaker may not be suitable for tube amps with high output impedance, due to other reasons ........ See, JA1's comments in measurements section :-) .........

Bogolu Haranath's picture

Tube aficionados would say, 'tube sound can only be imitated, but never duplicated' :-) ........

Bogolu Haranath's picture

You may already know this ....... Wikipedia has information about 'Impedance matching' ....... There is also some information about 'conjugate load matching' in that article :-) ........

Bogolu Haranath's picture

BTW ...... Each 8c speaker has a total of 1000 watts of amplifier power ....... How much do you think D’Agostino would charge for 1000 WPC amplifier(s)? :-) ........

Bogolu Haranath's picture

You can read about both feedback as well as feed-forward designs, in Benchmark media blog ....... They explain what each technique can do ....... Benchmark uses both feedback as well as feed-forward techniques in their AHB2 and HPA4 :-) .........

Ortofan's picture

... appropriate design approach?
Using negative feedback, feed-forward - or no feedback?

Bogolu Haranath's picture

I don't know ...... Most of the amplifier designers seem to use negative feedback only .... and they choose variable output impedances of the amplifiers with variable amount of harmonic distortions ...... Benchmark chose both feedback and feed-forward techniques for their amplifiers ....... Benchmark chose to have the lowest distortion possible ........ That the reason why their products have around 20 Bit resolutions :-) .........

Bogolu Haranath's picture

JVS listens to 24 Bit music with his reference amps which have about 12 to 13 Bit resolution :-) .......

Bogolu Haranath's picture

Did you do 'water-boarding' for you know who before, Ortofan? :-) ........

Bogolu Haranath's picture

One good example of how negative feedback affects the distortion levels, see the measurements of CH Precision integrated amp ....... CH precision is one of the companies that offers user adjustable negative feedback for their power amps and integrated amps ........ Another company is Cary Audio ........ Carver also offers that adjustable controls :-) .........

Bogolu Haranath's picture

I suppose, even an active speaker designer could offer user adjustable negative feedback controls, and make the active speaker sound like it is being driven by a tube amp ...... You could talk those guys at D&D and ask them whether it is possible, Ortofan :-) ........

John Atkinson's picture
Ortofan wrote:
1) the manner in which amplifier performance depends upon how much negative feedback the designer chose to use.

The problem with negative feedback is that with a typical amplifier circuit's limited open-loop bandwidth, there is insufficient gain margin in the top audio octaves for non-linear behavior to be suppressed. A better strategy is to design a circuit that has a wide open-loop bandwidth and as fundamentally linear as it can be before feedback is applied.

And no, that isn't easy. Bob Cordell discusses this subject in the second edition of his book Designing Audio Power Amplifiers, which I review in the June issue of Stereophile.

John Atkinson
Technical Editor, Stereophile

Bogolu Haranath's picture

Does Bob Cordell advocate for 'feed forward' amplifier output, which Benchmark AHB2 uses? ....... I think the new models of Chord amplifiers are adapting that technology :-) .......

Ortofan's picture

... the "better strategy" of wide open-loop bandwidth?

Bogolu Haranath's picture

If you can come up with a way you can fit tube amps inside those 8c boxes, you could become a multi-millionaire, Ortofan :-) .......

Bogolu Haranath's picture

Graphene, an allotrope of carbon is supposed to be 200 times stronger than steel ...... It is also very light weight, lighter than paper ....... Manufacturers like Joseph Audio and Magico use Graphene cones ........ Graphene is expensive :-) .........

Bogolu Haranath's picture

Graphene is supposed to be stronger than diamond! ........ Perfectly made Graphene is supposed to handle 14 million pounds per square inch pressure before it breaks! :-) .........

Bogolu Haranath's picture

Tidal Akira ($215k, reviewed by JA1) uses diamond diaphragm tweeters and midrange drivers .... (JA1 is saving money to buy Tidal Akira) :-) .......

tonykaz's picture

$12,500 might seem a little pricy but it has it's own Amplification and Room Setup Brain.

All European, clean appearance .

Reminds me of Genelec and Meridian.

$12 Grand will buy a full Schiit & Klipsch or get a person a good running start on a PS Audio rig but it won't move the needle for the seasoned Audiophile about to spend 12 grand on a phono cartridge.

D & D is a bonafide Engineering Company, kinda like Porsche Design on a much smaller scale, I think that they belong here.

Tony in Venice

Bogolu Haranath's picture

Boris Johnson cleaned up his hair-style after he became PM :-) ........

tonykaz's picture

He rides a bike to work, I'm told.

Tony in Venice

Bogolu Haranath's picture

So, Boris Johnson is a clean European who uses clean energy :-) ..........

tonykaz's picture

Hmm, lets hope

Tony in Venice

Bogolu Haranath's picture

Boris Johnson may even ride an e-Jaguar, which uses clean battery power :-) .......

Ortofan's picture

... https://autojosh.com/new-uk-prime-minister-boris-johnson-will-use-armoured-jaguar-xj-sentinel-as-official-car/

tonykaz's picture

I'm as clean and pure as the Wind driven Snow, by comparison.

my-good

Tony in Venice

Bogolu Haranath's picture

Mr.Tony, you could talk to GM and ask them to come-up with a battery powered 'Beast' :-) .......

tonykaz's picture

rather disappointing overall.

but...

GM makes money from Service with the Cars needing regular service being the overall formula.

Cars like Tesla don't have Service Intervals, they're built to run 1,000,000 Miles, they're like Locomotives. Where's GM's profit in that?, where's the Dealer's profit?, where's the service Tech's profit?, where's the aftermarket parts profit?

By the way, Henry Ford built the FMC around the one car lasting the rest of your life concept back in Ford's early days. Ford didn't change to the 11 year durable good concept until after WW11 and Henry was dead ( 1947 ).

Henry Ford invented the Tesla concept.

Tony in Venice

Bogolu Haranath's picture

Armored Jaguar pales in comparison to the armored 'Beast' :-) ........

Anton's picture

I hear that during his off time he drives a Lotus that can turn into a submarine.

Ortofan's picture

... playing on the ICE system:

https://www.youtube.com/watch?v=BTFfW29DP24

Bogolu Haranath's picture

The PM also drives an AMC Metador coupe, which can turn into a flying machine :-) .......

Bogolu Haranath's picture

And ..... The PM used his 'Bell Rocket Belt' for his Brexit :-) ......

tonykaz's picture

but...

I hav'ta suggest that folks have a look at "The Jazz Shepherd" YouTube dailys.

Fascinating content from a fascinating person, a supporter of our "Community"

Tony in Venice

Robocop's picture

I have a query with rear firing woofers. The substrate makeup of the wall these are facing must surely have a major effect on the sound alongside the increase in db's. I guess DSP can potentially overcome this. I'm thinking particularly where one has a brick wall, concrete, concrete block, clad plaster or wood in all its variations. Even plaster board on wood supports which is what I have is not a total rigid surface and potentially with loud music is moving physically in the non supported parts. I have never played with rear firing woofers to ever consider this aspect.

X