Dutch & Dutch 8c Speakers

[tmallin]

My reference to everyone agreeing that flat on axis speaker response and listening window response is correct is in terms of ANECHOIC measurements (or software like the Klippel which apparently can mimic anechoic measurements even though taken in a real room) of the type shown in spin tests.

I'm NOT talking about the usual in-room-point-a-mike-forward-toward-speakers-from-listening-position that I and many others do. Such measurements, if they show truly flat response on top indicate the speaker is way too bright/aggressive sounding in that room. And if the bass response is very flat in such a measurement, the speaker will sound thin. Thus the preference for target curves for such measurements to be a bit up in the bass and a bit down in the treble.

If you don't believe that anechoic speaker measurements on or near axis should be flat, then you are accepting a different standard for speakers than all other audio equipment in the chain. Flat anechoic response on or near axis should be the target.

How much the first reflection and sound power anechoic measurements should differ from flat is the area of dispute. And as I said, in a larger room where the first reflections are delayed by 10 ms or more from the direct sound, they may not matter subjectively nearly as much as in smaller rooms like mine where the first reflections are integrated into our perception of the tonality.

Perhaps I'm viewing things differently as I get older. I don't want to fiddle with frequency response so much from recording to recording. I want a set up which sounds great on most recordings with little fuss. I have no doubt that listening close up to BBC-type speakers toed in to face my ears in a well damped room--sounds great. I've used that set up many times in the past and it DOES sound great--most recently with the Harbeth M40.2 in my current room with the speakers only 55 inches from my ears. I think the "well damped" room part is more difficult to get correct in a small room like mine than in more capacious rooms, but you can get it right enough for great musical satisfaction.

I think people need to understand more about what the spin tests are and what they show. I suggest reading https://www.audioholics.com/loudspeaker-design/understanding-loudspeaker-measurements for an easy-to-follow explanation. The spin tests are meant to be done anechoically or with software (like the Klippel) which allows the measurements to mimic such conditions even though the testing is done in a real room.

These tests have much more universal application than JA's Stereophile measurements, for example. JA's measurements may only be relevant in his listening room. Other rooms obviously show much different measurements taken with JA's same technique as is sometimes evident when he compares measurements taken in his room with measurements of the same speaker taken in the reviewer's room.

 

[tmallin]

RE: Concentric driver speakers vs. line sources and others:

A concentric point source box speaker does not interact with room modes in the bass any less than other types of box speakers. Similarly, your listening position will alter what you hear in the bass the same way with point source speakers as with other speakers.

I've used the Gradient Revolution Active and Gradient 1.4 concentric tweeter/mid speakers. Yes, there is a bit more focus possible, but with the additional image focus comes a bit less large image height. The Revolution was more problematic with this than the 1.4 since the 1.4 allows you to alter the angle of the mid/tweet sphere to point up a bit. If you aim the spheres at your ears and sit down low enough so that your head is not much above the sphere, the images and stage will appear right in front of you, not down, and there is sufficient image height. Other speakers, like Harbeth M40x or my current D&D 8c will give you larger/taller images, though, without any noticeable image smear.

I have listened to Tannoy Dual Concentrics. With most of the models, listened to from close up, since the center of the concentric array is actually pretty low off the floor, you will tend to look "down" on the images and stage a bit. Of course, you could always put the speakers up on stands of some sort to move the concentric driver higher up. That will reduce the bass response somewhat, however, and since I find the tweeters of the Dual Concentrics I've heard to be a bit on the hot side, that's not a move in the right direction. EQ might help.

If by line source you mean a continuous really tall ribbon or some such driver, yes it can create enormous image height. The precedence effect will still keep the focal point of images straight in front of you, but in my experience there will still be some vertical stretching of images. Whether you think this sounds more life-like large or exaggerated is a judgement call, I think.

Some line sources are constructed of closely spaced small drivers. Whether there are audibly obnoxious interference effects from such an array depends not only on how well the manufacturer has managed to integrate the multiple drivers covering the same range into a coherent sounding whole, but also on whether you stay absolutely still in your listening position and don't try to hear the interference effects by bobbing your head up and down or side to side, and--more than anything else--how close you sit. The closer you sit, the more likely you are to hear out individual drivers and hear interference effects.

 

[tmallin]

I agree that envelopment is a holy grail for home listening as well as concert hall listening. We two-channel guys are at a disadvantage since one "easy" way to gain some envelopment is to put real ambient sound sources around and behind you as with a surround sound speaker set up. How best to gain envelopment with just two speakers is the challenge.

Putting reflective boards near the sides of your speakers to angle the side wall reflection down the room past the listening position may add much needed delay to the side wall reflections, but how can this be done with floor reflections, ceiling reflections, and reflections from the wall behind the speakers? Getting a 10 ms delay may be easy for the side walls and wall behind the speakers with a large room. And if you have a 10 foot ceiling you can do it for the ceiling reflection. But the floor reflection remains a problem. Perhaps absorption is the best you can do for that reflection and that's what I do.

The reflective path to my ceiling is not quite 10 ms delayed compared to the direct sound in my room, I think. It is close, though, since the ceiling is 103" high and the tweeters are only about 40.5" above the floor, and there is the angle of reflection to add additional distance/delay. In any event, even with this much delay, I notice a rather obvious increase in envelopment now that I finally figured out how to mount foam batts on my ceiling in a non-damaging-to-the-ceiling manner. I now have all the mirror-like surface reflections of my room treated with 4-inch acoustic foam except for the wall behind the listening position which is treated with P.I. Audio Group's AQD diffusers.

While some recordings now sound quite immersive, others don't. Recordings where phase is obviously being doctored to enhance spaciousness (usually electronic or pop music) can sound quite impressively immersive. Some classical music recordings are also excellent at this.

But this is the primary area where I crave more from two-channel listening. I remember fondly the incredible immersion possible when intra-aural crosstalk cancellation circuitry was in play with my old TacT processor's XTC circuit (or Carver Sonic Hologram, or Hughes SRS Processor, or the inconvenient physical barrier method, for that matter). Just please eliminate the phasey "tugging at my ears" effect and that would be a huge advance in two-channel listening for classical music, I think. The simplest solution may be a pair of real rear speakers with adjustable volume control a la the old Hafler Dynaquad system.

But I wonder how well this can or could be done with just two speakers these days with more advanced processing. I plan to hear the Polk Audio L800 system soon since it has garnered some very positive reviews for its envelopment, particularly for classical music if you sit in the sweet spot (duh). The L800 revives the old Polk SDA technology in a speaker which is said to work well close to the back and side walls, if that can be believed. With the old TacT XTC processor, the best speaker placement by far was far out into the room with the speakers separated side to side by at most two feet with no toe in. Quite an interesting setup for the Gradient 1.5 Helsinki's I was using at the time and the imaging and staging were incredible, totally belying the physically almost touching lack of stereo separation of the actual speakers. I would have kept using it but for the phasey ear-tugging effect.

 

[tmallin]

Some people seem to like the "light up the room effects" of dipoles, bipoles, and omnis.

Other people, like me, cringe at the thought of all those highs bouncing off room surfaces waaaay too early to add any semblance of realism.

I think I may part company with some other reviewers on the issue of how much dispersion in the mids and lower highs there should be to avoid a speaker sounding too dead. But then I think a lot of that potential disagreement may come down to how small our respective listening rooms are. My room is so small and the walls necessarily so close to the speakers that I like even less room reflections than others do.

Even with direct radiator speakers, the idea of intentionally flinging bright sound upward to add spaciousness has been around a long time. For example, in the early 1970s, Winslow Burhoe's EPI/Epicure speaker company marketed the EPI 201/Epicure 20/20+ which had one of his modules facing forward with the other pointing a bit forward, but mostly straight up. Both modules were hot in the highs if heard on axis (as was his most famous speaker the EPI 100 which had but a single module). But placed as intended, against the wall behind it and on the floor, a seated listener was not on axis with either module's tweeter. Talk about a wide open sounding room lighter upper! Not focused very well at all, of course, but it sounded big, open, and clear. A pair on the used market today can usually be had for $500 or less, e.g.: https://www.ebay.com/itm/Epicure-Mo...602603?hash=item2f48d85f2b:g:tC0AAOSwHlhfQhl1

 

[tmallin]

RE why many speaker reviews these days don't stress the requirement of sitting in one centered position for best spatial effects:

I suspect that there are many reasons this in not emphasized or even mentioned in many reviews. One reason is that is may be assumed. I think this is the least likely reason, however.

Another reason is that the assumption today is that speakers should sound good in stereo from several seats. The reason for this is that solo listening, especially from a fixed seat, is regarded as politically incorrect these days. We shouldn't advertise that audiophiles are anti-social or don't listen in groups. We don't even want to mention that, for group listening, listening in a train arrangement (one listener behind another) is better than side by side since a train formation just makes us look weird.

The most likely reason it's not mentioned, I think, is the assumption that your stereo does double duty as a home theater. As such, group listening is assumed even for music listening. The whole idea behind the listening window measurement in the spinorama is to show how flat and smooth the speaker response is over the plus or minus 30 degree horizontal and plus or minus 10 degree vertical angular window. That is assumed broad enough to cover two rows of three seats across from a reasonable home theater viewing/listening distance.

But here's a very important caveat: even from a single-person close up seat, the dispersion of the speaker matters, at least for the discriminating listener. Even when I listen in my room from 55 inches from the drivers, I can clearly hear the effects of room treatments on what I hear from the sweet spot. Factors that make the effect of room treatments less obvious are listening close up, toeing the speakers in to face my ears, and narrower horizontal and vertical high frequency dispersion of the speakers. But even optimizing those factors, the presence and nature of the room treatment still clearly affects what I hear from the sweet spot.

Perhaps from a listening distance of 20 inches or so room treatment wouldn't matter much. I really have not had a set up where that is practical. Somewhere online I did see a picture of a D&D 8c set up with an iMac like I have on a small desk with the speakers just flanking the desk and positioned forward of the screen. That would be a full system. Roon, Qobuz, Tidal, internet radio, and music files would run on the iMac with the D&Ds providing the rest of the system. My current computer desk is too big and the side walls are too close for such a set up. This would work well with a small computer desk centered on a long uninterrupted wall. Perhaps room treatments would not be necessary or even audible for such a listening system. But that would, of course, be an extremely unusual set up, the closest thing to headphone listening one could devise for a practical speaker listening set up.

 

[tmallin]

Re: Some people suggest adding additional drivers to provide delayed reflections to improve the spatial qualities of reproduced music:

All the more reason for not adding extra highs to either the direct or reflected sound at the level of the home reproduction system. If you want to emulate the frequency balance of what an audience at a live unamplified concert hall performance would hear, that balance is relatively flat in terms of initial transients even at considerable distance, but the hall sound rolls off considerably starting around 2 - 4 kHz due to air path absorption, concert hall absorption from seats, people, carpet, etc., and reflections from less than perfect structural reflectors.

The microphones used to make commercial recordings are usually up high and closer to the instruments, both of which factors tend to exaggerate the amount of high frequencies captured by the microphones. Bass is stronger near the floor and many instruments project high frequencies up, the air path losses are less, and there are no reflective losses. Granted, the recording and mastering engineers can correct this through equalization back to something more plausibly like what an audience would have heard. But the prevalence of classical recordings which sound both too bass thin and too high frequency bright through most speakers which measure reasonably flat anechoically suggests that the recording studio compensation is not complete enough in many cases.

 

[tmallin]

The Active Room Management (ARM) of the D&Ds as I have it set, affects only the bass end below 200 Hz or so. You dial in the actual distances of the rear subs from the wall behind them and the side wall. This is intended to compensate for room boundary Allison effects.

The ARM also has settings for the subwoofer level (100 Hz and below), the range from 100 Hz to 1200 Hz (controlling the range of the bass/midrange driver on the front), and the tweeter (1200 Hz on up). I have the midrange and tweeter controls set at "flat." I reduce the subwoofer level by 2.5 dB so that 20 Hz is 6 dB above 1 kHz. Otherwise, the 8c's produce a lot of low bass in my room, even for my tastes.

Besides the ARM adjustments, 24 bands of parametric EQ are also available for each channel. I currently use three bands, one to boost the 200 Hz region to counter the "usual floor dip," one to decrease the area around 400 Hz to counter the resonance shown in the spin tests, and a third to raise the 20 Hz level to follow the bass uplift curve so that it continues below 30 Hz.

The amount of bass lift I have dialed in was adjusted through a lot of listening to both pink noise and music recordings, both classical and jazz with walking bass lines. I want jazz bass to be clearly audible, but not heavy and I want to very clearly hear the tune on string and electric bass scales. Any more lift at 20 Hz and bass drums, organ, and hall sound starts to get a bit "billowy" or "fwoomy" and pink noise starts to develop a "tone" at the bottom. Any less, and I start to lose the fantastic bass foundation I now have. I may want to experiment with boosting the 100 to 200 Hz region a bit for more low brass power, in effect starting the bass lift at 200 Hz rather than 150 Hz.

Any more treble, even 0.5 dB lift of overall tweeter level through the ARM control, starts to brighten things up a bit too much on a wide range of material. This treble balance seems "just right" to me on most material in this room through these speakers with my present room treatment.

Eventually I'll try REW equalization with the 8c's but they sound so fine this way I'm not in any hurry.

 

[tmallin]

Re: why my preferred measured response of the D&D 8c's does not have a smooth roll off from bass to treble:

Yes, Mitch Barnett likes the way the 8c's sounded in his room, as I do. And I certainly agree with Mitch that the 8c's are the first speakers where I really don't need or want to turn down the highs on most material I listen to.

Note that Mitch's listening material is not similar to mine at all. His choices are much more rock/pop oriented and such recordings on average certainly are balanced much hotter than my much more eclectic mix of classical, jazz, electronic, folk, world, etc. Rock and pop recordings in general are quite compressed and nastily bright. An extra rolloff might well be preferred for a constant diet of that.

Back a decade or more ago when I had the TacT RCS 2.2XP, I experimented with target curves which sloped constantly from 20 Hz down to 20 kHz in various amounts. It didn't take me long to conclude that I preferred targets where the midband was relatively flat. Back then my listening was more classical-music-weighted than it is today.

 

[tmallin]

Re: Breaking the Toole/Olive "Audio Circle of Confusion":

Sigh. On the one hand, you have hopelessly idealistic/impractical me arguing for some sort of agreement on the part of mastering engineers about the equipment and set up they use to finalize how their recordings sound, something which could then be mimicked at home to a greater or lesser extent in an attempt to hear recordings as they are meant to be heard. This is a clear but commercially impractical and very expensive for the home listener path forward for hearing perhaps 99+% of available newly released commercial recordings at their best.

On the other hand, you have others basically saying that all you should tune your system to is the sound of some very small number of commercial classical recordings which we feel to best represent the sound of real acoustic music heard in the original acoustic space. For the rest of recorded music, just reproduce it in whatever way that pleases you, because such recordings are, after all, kind of beneath our elevated standards and thus are not worth worrying about too much. Tune to the best, and as for the 99+% of the rest, let the chips fall where they may. This view, which seems to reject the relevance of all music other than classical, and of all recordings other than those made via "purist" methods, could reasonably be regarded as both curmudgeonly and elitist, not really fit for discussion in current politically correct circles, and one which would result in the reviewer being "cancelled" by modern culture.

Not a very good choice of options. This is why I, having these days rather eclectic tastes in music listening, have tended to waffle. Sometimes I lean in favor of system components and set up which are pleasing to me in my listening room on a wide variety of commercial recordings. Other times I stress hearing what is actually on the recording, warts and all.

I did not mean to imply that the method of focusing on maximizing the realism of a few best recordings will lead to terrible results for other recordings. There is much to like about semi-anechoic playback on a lot of recordings. You just have to be prepared to find at least some recordings rather lifeless without moving further back in your room out of the sweet spot and you will have to equalize a lot more recordings to bring bass and especially treble into more realistic or even tolerable proportions.

The other approach, trying to find components and a set up which creates maximally pleasurable results from a wide range of recordings with minimal futzing from one recording the next once you settle on what, for you in your room, seems a happy medium, sacrifices the maximum degree of spatial and tonal realism attainable from any given recording, including the "best" recordings.

One good thing that might come from widespread agreement among pro-audio folks about the worthiness of the Toole/Olive "scientific research" is greater standardization at the recording end. Imagine if 90+% of recordings were mastered using speakers rating high on the spin test criteria. Home listeners would at least know what kind of speakers they should use to get closer to hearing what was intended in that big swath of recordings. Whether this is actually happening among pro-audio folks, I don't know.

 

[tmallin]

Re: View that the D&D 8c anechoic response is a bit down at the treble end, even on axis:

Yes, the shape of the anechoic response of the D&Ds on axis and in the listening window might be termed "relaxed" or "forgiving" because the highs are a couple of dB reduced from the midrange on axis and in the listening window response measurements. But that works out well with the great bulk of recordings since it seems to me that even most classical recordings are at least a bit "hot" in the highs as heard through most high-end speakers, if likeness to the real unamplified sound of instruments ("natural" sound) is the benchmark.

For those wanting more treble from the D&D 8c speakers, the entire range of the tweeter (1.2 kHz and up) is adjustable in 0.5 dB increments up or down with one finger touch for both speakers at once through the Web interface with the speakers on my iPad from the listening seat. No, that won't "fix" any dispersion characteristics you may disapprove of, but it will adjust the treble balance pretty easily to overall taste or for particular recordings. As I said, I prefer it at the factory setting of "0" dB. I've tried adding treble, but while even adding 0.5 dB wideband obviously is audible, even that much is too much for my tastes. Adding +2 dB is quite a change and I don't think most folks would find it more natural sounding on most recordings. But this adjustment IS there and easily engaged for those who want to experiment with a different overall balance. It is certainly easier to engage than on most speakers with pots or switches on the cabinet and the gradation of options is at least as fine as on other speakers with adjustable treble.

The mid-woofer (100 to 1200 Hz), and the sub-woofer (below 100 Hz) levels are similarly adjustable from the listening seat.

All these adjustments are in addition to 24 bands of parametric EQ per speaker which can be engaged to make narrower-band adjustments.

An addition to the D&D 8c firmware I hope D&D will consider is a digital domain simple bass and treble control. It would operate in steps no coarser than 0.5 dB at 20 Hz and 20 kHz. Ideally this would operate in Lanspeaker with a touch-screen bar-graph slider operable through my iPad since I, like many/most Roon and D&D users, use an iPad for system control anyway.

The options would be bass control only, treble control only, and a combination "tilt" control which operates both bass and treble at once. I would not want a 1 kHz pivot point for the controls such as some "tilt" controls (e.g., Quad preamp and Audient graphic equalizers) have implemented, however. I'd rather leave the heart of the midrange alone. I'd prefer inflection points around 200 Hz and 2 kHz, much like the tone controls implemented in the old Acoustic Research (integrated) Amplifier, graph shown here: http://www.classicspeakerpages.net/...nouncement_s/ar_amplifier_announcement_s.html

Toole, Robert Greene, and I all agree that the lack of very-easy-to-use bass and treble controls in modern audio systems is lamentable. Such a control would go a long way toward easily correcting a lot of what is wrong with many recordings, tonal balance-wise.

I realize that some components, such as the Dual Core and X4 produced by DSPeaker have something which approaches this type of control, but it's still not easy enough and the GUI is impossible to read more than a very few feet away.

There may be other software which includes such simple bass and treble controls but, if so, I'm not aware of it. I'm frankly surprised that Roon has not implemented this sort of control through its DSP function, but doing it at the speaker level would be even more ideal for many users.

 

[tmallin]

For some time Martijn Mensink has been urging me to try using REW to equalize the D&D 8c's. He told me I should try doing the corrections based on 1/24th octave smoothing or with no smoothing at all (instead of the 1/3-octave smoothing I used. He also suggested using spatial averaging of measurements. He thought I would appreciate how much tighter still the bass can become.

Now that I finally got my ceiling reflection points covered with absorbing foam (and what a nice improvement that was, as it has turned out!) the next step was to screw up my courage to try REW. I had no more excuses left.

Dutch & Dutch has written instructions (no video yet, as far as I can find) as to how to use REW to adjust the 8c's online here:

https://support.dutchdutch.com/rew/

I hesitated for so long because (1) the sound was already so excellent, (2) I had already manually adjusted the EQ for my room using my OmniMic V2 system, and (3) the directions for using REW discussed in most online video and print tutorials seemed pretty complex to me.

But, as it turned out, if you follow the D&D instructions exactly, it's not bad at all. The whole process is a lot simpler for the D&D 8c's than for most other speakers since D&D has integrated REW into the DSP of the speakers. IGNORE the tutorials elsewhere on the Web; they are way over-complicated for this application! D&D's instructions are correct enough. A few more screenshots would have helped even more, but it's really not all that hard. I followed the measurement instructions for a single-person listening area.

You will need to download the free REW program and you will need to have a measuring mike. But you don't need the specific microphone mentioned in the instructions. My OmniMic V2 worked fine; I suspect that most any USB-connected microphone with a calibration file would work about equally well. I used my Microsoft Surface Pro to run REW, just as I did with OmniMic.

I recommend saving all the measurements (all seven measurements plus the average of these measurements for each channel) to a folder, one folder for the left channel and one folder for the right channel. This might come in handy at a later date should you want to alter the Target Settings without making new measurements. At least save the average measurement curve for each channel.

I also would recommend taking a screen shot or photo of the screen shown in Figure 4 of D&D's instructions once you get that far in the process of applying REW to your first speaker. Yes, use right-click to copy the settings as the instructions say, but you may lose that copy if you happen to use your copy function for something else in the meantime and you may lose the copy if you shut down REW between your first and second speaker measurements. The screenshot or photo will allow you to make sure that you use the same settings for the second speaker.

One thing not mentioned in the instructions is how to set the low frequency slope. You can use 24 dB per octave as shown in Figure 4, but I ended up using 18 dB per octave since that target was a better match for the slope of the measured low-frequency roll off of the speakers in my room.

Another thing not mentioned in the instructions is how to change the Target Settings much later on if you decide to experiment with other Target Settings. I suspect that this is as simple as loading the measured-and-saved average response for each speaker, adjusting the Target Settings to taste, then reapplying the altered Target Settings to the speakers under the Filter Tasks panel.

One potential issue: I found that the SPL of the test tones was a bit or more too high when setting the volume control at minus 30 dB as recommended in the instructions for my speakers' serial numbers. I found that to get an average target level around 80 dB after measurement, the test sweep level for measurements needed to be set at about minus 38 dB. This was with firmware version r1.4.57. The recent update to firmware version r1.4.65 may have corrected this since the notes for that version say that it "fixed an issue that caused REW to play louder than intended." I have not tried measuring since that firmware update.

So how does the result sound in comparison to my prior manual adjustments? The difference in measured response is subtle, but the effect doesn't seem so subtle. Another level of excellence is revealed. Smoother bass certainly, but things also seem more focused and 3D, and everything just seems cleaner-yet sounding. As Martijn once told me in an email, "The filter optimizer in REW works a treat."

I strongly recommend that owners of the D&D 8c speakers you give this a try. You may well be very pleasantly surprised.

 

[tmallin]

As I suspected above, re-doing your EQ settings via the REW integration with the Dutch & Dutch 8c is simple. Here are my step-by-step instructions. REW's own instructions define terms for this process and are also otherwise helpful and are found here:

EQ Window

It takes a lot longer to read my instructions below than to make the changes once you go through the process even once:

1. Open REW and expand it to full screen. Click on the Open tab near the top left.

2. Open the measurements you saved for the left channel speaker. There should be seven measurements plus the average. If you saved each measurement separately, including the average, all you need to open is the average measurement.

3. Highlight the average measurement in the left panel so that it appears in the main panel to the right. Click on the "EQ" button which appears above the main window near the top of the screen. Expand that window to full screen.

4. Expand the first tab of the panel to the right, make sure "Equaliser" is set to "Dutch & Dutch 8c".

5. Expand the second tab, "Target Settings", and manually make whatever adjustments you think you want to try. Leave the "Target type" set to "Full range speaker". Also leave the "Target Level" set the way it was originally. I suggest possible adjustments to "LF Cutoff", "LF Slope", "LF Rise Start", "LF Rise End", and "LF Rise Slope".

6. Unless you really want to decrease the high frequencies of the 8c response, make sure the "HF Fall Slope" is set to 0.0. This probably doesn't matter if you are only equalizing the bass, but this is double protection that no changes will be made to the speaker's built-in high frequency response.

7. Once you have set the "Target Settings" the way you want them, right-click anywhere in that pane and choose "Copy Target Settings". (You will need these for the right speaker later in the process.) Also take a photo of the screen to make sure you can recall these settings if you lose the copy from your computer's clipboard.

8. Expand the third tab of the right panel, "Filter Tasks". Set "Match Range" to the desired range. 20 Hz is the default lower end of the range and you should pick the upper range based on eyeballing your saved average measurement and set it high enough so that it will apply to the major wiggles to the left side of the graph. I'd suggest going up high enough to flatten the 400 Hz bump that several reviewers (including me) find in the response. But, as I said, pick the frequency based on eyeballing your own in-room average measurement.

9. Set "Individual Max Boost" to 6 dB. Leave "Overall Max Boost" at 0 dB. Make sure the "Allow narrow filters below 200 Hz" box is checked.

10. Click on "Match response to target." The system will calculate for a few seconds.

11. Then choose "Send filters to speaker or group." From the drop-down menu which appears, choose the serial number or name of your left speaker and click OK. You should then receive an acknowledgement that the filters have been sent. (If you get a message saying no Dutch & Dutch speakers were found on the network, select "Scan network for speakers" and pick the left speaker.)

12. Repeat steps 2. through 11. for the right speaker and you are done. In step 2, you need to open the average measurement for the right speaker, not the left speaker.

You can do all this while listening to your system so you can then compare the new settings with the old on known material. You can do more iterations until you think it sounds "right" for your room, set-up, and ears.

I find that I am very sensitive to the "proper" level of low frequency boost when equalizing the low end to remove peaks without filling in dips, which is what REW and many other good equalization schemes (like the DSPeaker X4 and Dual Core) do. Once the peaks in response caused by your room modes are lopped off, the bass can sound "thin" without the proper amount of smoothly applied bass boost to compensate for the removal of the peaks. For example, with a low frequency rise set to start at 200 Hz and proceed down to 20 Hz, I can clearly hear the difference between a rise of 1.8 dB per octave and 2.0 dB per octave and can hear that 1.9 dB per octave is midway between these rise settings. To my ears, in my room, I am finding the 1.8 dB per octave setting to be ideal.

Have fun!

 

[tmallin]

I've recently been reading texts on loudspeaker design and use, both from a professional and consumer viewpoint. So far I've read the books by Toole, Colloms, and, most recently Newell & Holland "Loudspeakers for Music Recording & Reproduction" (2d edition). That last book is, in my opinion, the most helpful I've seen in terms of explaining, in a not-too-technical manner, the why's and wherefores of loudspeaker design and usage, as well as giving practical advice for both pro-audio and consumer users. Newell & Holland also seem to best balance objective measurement data with subjective quality evaluation of loudspeakers.

One of the things I've concluded as a result of my now-extended use of the D&D 8c speakers is that fully active design with DSPed crossovers and thus true time alignment and smooth frequency response is the way to go. As Newell & Holland note, however, it took a long time for even pro-audio users to reach this conclusion despite overwhelming evidence for it as compared to passive speakers. And with consumer audiophiles, there is still huge resistance to this approach:

"Perhaps the only real obstacle to the general acceptance of the superiority of active crossovers and multi-amplification in the world of domestic hi-fi is the fact that so many hi-fi enthusiasts want to select their own favourite amplifiers and loudspeakers as separate items. This could have more to do with human psychology rather than audio engineering, but of course, choosing your own system with future upgrades in mind, as extra money becomes available, is a fun part of building up a hi-fi system, and may still be a necessity in some smaller studios. However, it must be stated that in many less-demanding circumstances than studio-monitor loudspeakers, passive crossovers obviously have their appropriate applications, but the [below] lists highlight the benefits of active designs where the highest system performance levels are required. Clearly, the advantages of active, low-level crossovers completely eclipse those of passive, high-level crossovers, yet it was only around the late 1980s that dedicated active crossovers began to be seriously used on large-scale [professional studio] monitor systems. Prior art used standard electronic crossovers, and frequently ignored the effects of group delays, leading to some very poor transient responses. This lack of attention to detail led to some delay in the acceptance of totally active designs because of the uncertainty about the perceived benefits. The use of crossovers with fixed slopes on all the filter bands tended to necessitate the use of multi-band equalisers, many of which were of dubious sonic quality. Many people still expected to be able to mix and match stock equipment, so it took some time before people generally began to accept the need to buy a specific crossover with a monitor system, and which would be relatively useless in any other application. There was a lot of resistance to not being able to choose one’s own favourite amplifiers to use with whichever loudspeaker system, with each piece of equipment expected to function as a ‘stand-alone’ device. In fact, it took a long time before self-powered, actively crossed-over small monitors could establish their place in studio use, but domestic resistance to their acceptance has been even more pronounced. Established practices and customer expectations can die hard, and they can be remarkably difficult to change, even in the face of clearly superior technology."

So, exactly what are the advantages of active speakers? Here's Newell & Holland again:

"5.6 Active versus Passive Crossovers For high-quality loudspeaker applications, the consensus is almost universally in favour of active crossovers for professional use. In the case of analogue devices, by virtue of their feedback loops they can remain remarkably stable over very many years, and complex filter shapes can be devised without any of the power loss that blights high-order, high-level passive crossovers. Conversely, passive crossovers rely entirely on the long-term stability of each component part for their overall stability, which is not easy to achieve when the low impedances of loudspeaker circuits call for high-value capacitors – in terms of both capacitance and working voltage – which in turn call for capacitors of types which may not be able to provide good, long-term stability. Complex filter shapes may need to use many components, wasting large proportions of amplifier output. If large electrolytic capacitors are needed, their stability can be questionable, but, if the much larger, solid dielectric capacitors are used, their physical construction and large size can lead to them having considerable unwanted inductance, which can upset the crossover operation. They also tend to be very expensive.

"Active filters are free from these problems, and in fact require no inductors at all. What is more, if state-variable filters are used, any drift which does occur can reflect equally in both halves of the crossover response, only slightly varying the crossover frequency as opposed to opening a gap or causing an overlap. The overall frequency response of the system is therefore unlikely to be affected. There can be no equivalent to this type of stability or self-correction with passive crossovers, and neither can passive crossovers easily compensate for group delays. Figure 5.22 shows a passive all-pass circuit (an analogue delay circuit) of a type which has been applied commercially to loudspeaker systems, but this type of circuitry between an amplifier and a loudspeaker can again cause as many problems as it solves – if not more!

"The list of advantages in favour of active crossovers and multi-amplification is impressive:

"1. Loudspeaker drive units of different sensitivities may be used in one system without the need for lossy resistive networks or transformers. This can be advantageous because drive units of sonic compatibility may be electronically incompatible in passive systems.
2. Distortion due to overload in any one band remains captive within that band, and cannot affect any of the other drivers.
3. Occasional low-frequency overloads do not pass distortion products into the high-frequency drivers, and instead of being objectionable may, if slight, be inaudible.
4. Amplifier power and distortion characteristics can be optimally matched to the drive-unit sensitivities and frequency ranges.
5. Driver protection, if required, can be precisely tailored to the needs of each driver.
6. Complex frequency-response curves can easily be realised in the electronics to deliver flat (or as required) acoustic responses in front of the loudspeakers. Driver irregularities can, except if too sharp, be easily regularised.
7. There are no complex load impedances as found in passive crossovers, making amplifier performance (and the whole system performance) more dynamically predictable.
8. System intermodulation distortion can be significantly reduced.
9. Cable problems can be dramatically reduced.
10. If mild low-frequency clipping or limiting can be tolerated, much higher SPLs can be generated from the same drive units (vis-à-vis their use in passive systems) without subjective quality impairment. (See (2) and (3) above.)
11. Modelling of thermal time constants can be incorporated into the drive amplifiers, helping to compensate for thermal compression in the drive units, although they cannot totally eliminate its effects.
12. Low source-impedances at the amplifier outputs can damp out-of-band resonances in drive units, which otherwise may be uncontrolled due to the passive crossover effectively buffering them away from the amplifier.
13. Drive units are essentially voltage-controlled, which means that when coupled directly to a power amplifier (most of which act like voltage sources) they can be more optimally driven than when impedances are placed between the source and load, such as by passive crossover components. [When ‘seen’ from the point of view of a voice coil, the crossover components represent an irregularity in the amplifier output impedance.]
14. Direct connection of the amplifier and loudspeaker is a useful distortion-reducing system. It can eliminate the undesirable currents which can often flow in complex passive crossovers.
15. Higher-order filter slopes can easily be achieved without loss of system efficiency.
16. Low-frequency cabinet/driver alignments can be made possible which, by passive means, would not be feasible.
17. Drive-unit production tolerances can easily be trimmed out.
18. Driver-ageing drift can often be trimmed out.
19. Subjectively, clarity and dynamic range are widely considered to be better on an active system compared to the passive equivalent (i.e. same box, same drive units).
20. Out-of-band filters can easily be accommodated, if required.
21. Amplifier design may be able to be simplified, sometimes to sonic benefit.
22. In passive loudspeakers used at high levels, voice-coil heating will change the impedance of the drive units, which in turn will affect the crossover termination. Crossover frequencies, as well as levels, may dynamically shift. Actively crossed-over loudspeakers are almost immune to such crossover frequency changes.
23. Problems of inductor siting (to minimise interaction with drive-unit voice coils) do not occur.
24. Active systems have the potential for the relatively simple application of motional feedback, which may come more into vogue as time passes."

 

[tmallin]

Continuing from Newell & Holland's discussion:

"Conversely, the list of benefits for the use of passive, high-level crossovers for studio monitors would typically consist of:

"1. Reduced cost? Not necessarily, because several limited-bandwidth amplifiers may be cheaper to produce than one large amplifier capable of driving complex loads. What is more, the passive crossovers for professional use may require expensive components to ensure good sound quality and long-term stability.
2. It is often said that passive crossovers are less prone to being misadjusted by misinformed users, who think that crossovers are some sort of ‘adjust to taste’ tone controls, but many passive systems have a tendency to misadjust themselves with age and power levels.
3. Simplicity? Not necessarily, because very high-quality crossovers can be fiendishly complicated to implement, especially in large, high-output systems.
4. Ruggedness? No, because a change in characteristics of almost any component of the loudspeaker system can unbalance the crossover.

"In 2003 and 2004, Alex Campbell worked on a performance comparison between active and passively crossed-over domestic loudspeaker systems at the Institute of Sound and Vibration Research (ISVR) in the UK. He had been allocated an identical amount of money to spend on each design, and his findings were presented to an international conference of the UK’s Institute of Acoustics.7 The subjective assessments, made under ISVR control and using a panel of 30 subjects, came out heavily in favour of the active design. The results are reproduced in Figure 5.23, with the ‘clarity’ and ‘fidelity’ ratings being strongly in favour of the active designs. In fact, the only tendency for the passive design to show a significantly more positive result than the active design was in ‘brightness’, although this could have been the result of higher non-linear distortion levels."
. . . .

"5.8 Digital Crossovers Since the advent of digital signal processors of high sonic quality, the need for resorting to signal arrival compensation methods, such as the stepping of the baffles, can be eliminated if digital crossovers are used. Nevertheless, the physical effect of the non-collocated drivers still cannot be corrected over a wide area. This aspect of loudspeaker performance is something that just has to be lived with. Also, it should always be borne in mind that everything about any digital circuitry in a monitor chain must be of equal or higher resolution than the signal chain being monitored. Sonically, in terms of ‘hi-end’ hi-fi or high-resolution studio monitoring, the highest fidelity can only be perceived if the sample rate and bit rate used in the crossover circuitry at least equals that of the recording medium, or exceeds the resolution of the ear such that it can produce no audible artefacts. It may be difficult to hear the difference between a 20-bit/96-kHz recording and a 24-bit/192-kHz recording, for example, when listening to a crossover based on 16-bit/48-kHz processing. Furthermore, even if a crossover has internal processing which seems higher than necessary, the resultant output after signal manipulation has taken place may not be as great as the marketing figures would suggest. Historically, if these were to be of the highest quality, the cost of using digital crossovers could be exorbitant, because using anything less than the finest converters would make a mockery of trying to monitor recordings made through the best A-to-D converters. This was often a deterrent in terms of choosing to use digital crossover systems, but since the publication of the first edition of this book, the cost of excellent digital signal processors has fallen considerably. As time passes, the use of these devices as crossovers has become more commonplace in professional loudspeaker systems, although they have not been readily accepted in the world of domestic hi-fi. Digital signal processing is particularly attractive for studio monitoring because of the easy implementation of almost any amplitude response, phase response, signal delay, driver compensation, and even room-boundary-loading compensation, which can all help with the optimisation of responses in different working environments. A significant problem, however, is concerned with the converters. As the finest amplifiers are still of the analogue variety, D-to-A (digital-to-analogue) conversion must take place at each output of the crossover, and, for either professional monitoring or ‘high-end’ high fidelity, these crossover D to As must be of as high or higher resolution than any other part of the signal chain. When a signal processor is used as a crossover, it becomes a simple task to use the same device for delay compensation, out-of-band protection, over-excursion protection, boundary-loading compensation, and the control of other parameters, which effectively renders the signal processor to be an overall system controller rather than just being a crossover. Despite the relative simplicity and robustness of analogue crossovers, digital signal processors offer a wide range of possibilities both as crossovers and for the trimming of the overall system response."

I note that other than the possible loss of resolution from hi-res digital material downsampled to fit the performance of onboard DACs in the active speakers, the authors seem to acknowledge that digital crossovers are the way to go, now than high performance DACs are fairly inexpensive. Other sections of the book review the overwhelming technical superiority of the quality of closed box bass to ported or open baffle designs.

The D&D 8c's thus seem to check most if not all the right boxes.

But, then, there is this one concluding sobering thought: Newell and Holland admit that:

"A pair of ESLs from the 1950s can still put many of the latest loudspeakers to shame in terms of low colouration, low distortion, transient response, frequency-response flatness, and, perhaps most of all, perceived sound quality."

And:

"The authors of this book have, for decades, used full-range electrostatic loudspeakers as benchmarks against which to judge other loudspeakers, both objectively and subjectively. This is not to say that they cannot be surpassed on individual aspects of their performance, but their global performance is hard to beat."

There are in fact, several places in their text where Newell & Holland extol the virtues of Quad 63-type passive electrostatic speakers. Thus, the authors believe that passive speakers of a non-cone/dome/horn variety still have much to offer in terms of subjective listening quality even in the face of more modern actively powered and DSPed cone/dome/horn speakers with their "clearly superior technology."

 

[TerryM]

Thanks for the write up on your experience with the 8cs. I've owned a pair since August 2019. I agree with your assessment so I won't reiterate my findings. I really enjoy these speakers for the same reasons you do.

I did want to add one suggestion for those who like remotes. Take a look at the rooDIAL and roo6D extensions that allow for using a Microsoft Dial as a remote or a Space MouseWireless from 3dConnexion. I am not affiliated with the product, only a happy user.
roodial

 

[PCMusic]

Hi there. My nickname is PCMusic. I enjoyed reading the entire thread. Very informative. I wrote this review of 8c, and share many of your thoughts:

Apartment Audiophile: A ‘subjective’ review of Dutch & Dutch (D&D) 8c speakers for pop, electronic and rock music fans

This post is an ordinary audiophile’s ‘personal’ review of adapting Dutch & Dutch 8c's to enjoy pop, electronic and rock music in a slightly treated small apartment. Let me be frank: I am an ‘ordinary’ audiophile who has not done full room treatment (yet) due to domestic constraints and busy ...

audiophilestyle.com

If you are ok, I wonder if I could discuss something with you directly through personal message. I don't know how to do this though.

 

[tmallin]

I'm somewhat embarrassed to admit it, but until now I've never really tried listening to my system using the volume control built into the speakers to control the system volume. I suggest you try it. You may be pleasantly surprised. I believe that, as good as the Lumin Leedh volume control is, keeping the digital signal at full level until it reaches the speakers may be better sounding yet. Even less high frequency fizz, more stable imaging, and greater depth of field. I suggest other owners try it and see what you think.

With an all-analog system, from a gain structure viewpoint, to maximize signal to noise plus distortion, this is the ideal way to control system volume--leaving all the upstream components at full volume (assuming the output stage of any upstream component is not overloading any downstream component), but it's not usually practical since most amps lack a volume control or, if they have one, it's not usually handy to reach from the listening seat or is not remote controllable. Perhaps the same rules apply in an all-digital signal path. It seems to sound that way so far.

Of course, this requires that you leave the speakers connected to the internet home network all the time, but I was doing that anyway.

First, set the volume of the speakers at minus 40 dB. Then disable all upstream volume controls in your system or set them for 100% output. For me that just meant turning off the Lumin's volume control and leaving it at 100% of its full output level. If you are using Roon, also make sure that in the Roon device setup that your streamer is set to fixed volume. This should produce no higher than a fairly low level sound from the speakers on all program material. That level is good for background listening outside the room at my computer desk. You can turn it up from there. On some very quiet programs, like certain Sheffield and Reference programs, you may have to boost it to minus 5 or even 0 to get a nice loud level in the room.

Using my iPad Pro controller, I use the split-screen mode to let the D&D lanspeaker app occupy 1/4 of the screen with the Roon or Lumin screen occupying the rest. Volume can't be changed quickly using the D&D volume control (at least I haven't figured out how to do it quickly if it's possible) so you need to click the up or down volume buttons for each 1/2 dB change. Not a big deal since I find the useful volume range on most material is minus 30 dB to minus 10 dB, which is a total of 40 clicks up or down. Full mute is instantly available with one tap.

If you note that the tonal balance has suddenly changed and the speakers suddenly are considerably lacking in bass, you probably unintentionally engaged the Night Mode for the speakers. It is easy for your finger to slip while adjusting the volume up and down and accidentally touch the Night Mode icon. Check to see if the crescent moon icon on the Lanspeaker app screen is black rather than grayed out. If it is black, touch it to disengage Night Mode so that the crescent moon is again grayed out. The bass response bass should then be back to normal.

 

[tmallin]

Some reviewers think that the D&D 8c's horizontal dispersion is too limited in the highs and that in order to sound most like real life, speakers should have broader high frequency dispersion (like the BBC-inspired models) with wall reflections being then dealt with so as to be unproblematic. Such views on this dispersion and sidewall reflection issue are easy to explain for those whose listing rooms are considerably larger than mine. In some rooms, it's easy to get the speakers eight feet or more from the sidewalls. With that sort of distancing, the sidewall reflections are not that problematic with BBC-type speakers, as long as you listen in the relatively near field (say six feet or less from the speakers) and toe in the speakers to face your ears. And you can then eliminate any still-problematic sidewall reflections by attaching foam to the walls in the first-reflection areas.

Where speakers roll off starting lower than 2 to 3 kHz in the reverberant field, the speakers may well tend to sound a bit too muffled/airless in the top octaves in larger rooms. John Atkinson has often commented to this effect in his Stereophile reviews, for example.

Placing foam on the sidewalls works very well with BBC-type (e.g., Harbeth, Stirling, Spendor) speakers even in the smaller, narrower rooms (my current room is 132 inches wide, 161 inches long, and 103 inches high) I'm used to using as listening rooms, where it is difficult to get the speakers more than three to four feet from the sidewalls in any reasonable stereo set up. Such an arrangement has worked well with the Stirling LS3/6, Harbeth M40.2, and Gradient 1.4 speakers I've had in this room, among others.

But in my small listening room I've recently found that it is helpful to have more reverberant field roll off lower down in frequency to keep the sound even better focused and three-dimensional and eliminate excess brightness in the lower highs. The Dutch & Dutch 8c dispersion characteristics seem nearly ideal for this sort of small room, even though I have four-inch-thick foam panels at all the first reflection points, listen from 6.5 feet away and have the tweeters aimed at my ears. The "listening window" ( plus or minus 30 degrees horizontally and plus or minus 10 degrees vertically) dispersion pattern of the D&D speakers closely follows the on-axis pattern all the way out to 20 kHz. It is only further off axis that the D&Ds become more directional.

The felt strips of the LS3/5a or even a more extreme felt blanket over the entire baffle around all the drivers (e.g., the old Spica TC50) is primarily aimed at reducing diffraction effects from the speaker baffle's edges, not absorbing reflections from the walls.

If you don't like the look of foam panels on the wall (even cloth covered ones) you could try positioning some easily movable foam panels right next to the sides of the speakers--lean then against the speaker side walls if you want and have them extend out several inches beyond the front baffle to absorb the high frequencies before they ever hit the side walls. Foam panels in two-foot by four-foot size are ideal for either floor-standing or stand- mounted speakers. If you try that, I suggest using flat foam for this, rather than any with a sculpted pattern, or at least placing the flat side of the foam nearest the sides of the speakers. In my experiments, flat foam just sounds better in such a "near field" arrangement, but still not as fine as placing the foam on the sidewalls at the first reflection areas. Keeping the foam further away from the drivers allows the speakers to sound more "open" while still removing annoying brightness/distortion caused by "early" side wall reflections.

Another possibility is creating reflective panels made from plywood and angling these so that the sidewall reflections are aimed away from the listening position toward the area of the room behind the listening seat so that once those reflections finally reach the listener they are no longer "early" and are "late" enough to merely enhance spaciousness without causing audible problems. You'd have to judge whether such panels would also cause aesthetic problems in your room.

 

[tmallin]

I notice an audible improvement when damping the ceiling and floor first reflection areas with most speakers. This is quite a bit less obvious with my D&D 8c speakers than with any others I've had in this room or prior rooms. That's probably since the D&Ds are more directional than most and have a quite smooth vertical off-axis response. See Figure 4 at the following page from the Stereophile review: https://www.stereophile.com/content/dutch-dutch-8c-active-loudspeaker-system-measurements

Contrast this with Figure 6 from Stereophile's review of the Stirling LS3/6 on this page: https://www.stereophile.com/content/stirling-broadcast-bbc-ls36-loudspeaker-measurements

I think most people don't talk much about the vertical dispersion since damping the floor is left to carpeting in most cases (thick damping panels on the floor are a trip hazard) and ceiling dampening can be both more difficult to arrange and may seem less important since the ceiling is usually further away from the tweeter and other drivers than is the floor. In any event, in my experience, treating the side walls and wall behind the speakers always has had a more significant audible effect.

 

[Addicted to hifi]

I heard these Dutch 8c and was very impressed.