Dutch & Dutch 8c Speakers

Some seem to think that speakers which measure the way the D&D 8c does in spinorama tests (see the NextWeb review) will tend to sound "lifeless" in many home listening rooms.

The D&D 8c most definitely does NOT sound lifeless in my room to me. Frankly, I'm not really sure what "lifeless" in this context means. I THINK it means that a speaker which shows increasing directivity (in spinorama tests, for example) starting way down around 300 Hz or so, instead of up around 2 kHz or so, will tend to sound dull and "lifeless" in room.

But in the small narrow rooms I've listened in all my adult life I have yet to own a speaker which sounded lifeless as heard from the sweet spot. At shows and dealers I've heard a lot of JBLs, Revels, PSBs and others which follow more or less the "Toole School of Design" and I can't really say any of those have ever sounded lifeless to me in demonstrations. One exception: certain modern KEF models do seem reticent in the highs even from on axis.

In my rooms, when heard away from the sweet spot, the two speakers which sounded most rolled off/muffled in the highs were the Sanders 10C and the Janszen Valentina Active.

In my small narrow room, the dispersion of the D&D 8c seems ideal. Even with bare walls they sound very reasonably balanced and very listenable: no harsh brightness/brittleness, good staging and imaging, and you can play them loud without earbleed. Yes, they sound better yet to me with dispersive room treatment and yet better with absorptive room treatment near the speakers. But the differences are not huge as they are with most speakers, for example, the Harbeth M40.2s. And the tonal balance is reasonable anywhere in the room even with absorptive treatment near the speakers; I never get a "closed in" or "muffled" sonic impression when walking into the room with music playing or even from well outside the room as I'm sitting at my desk down the hall from my audio room now.
 
What is a "good" speaker? Here are a few possibilities suggested by Robert Greene on his forum:

1 A good speaker is one that in a plausible and obtainable
room situation--placement and room acoustics- makes recorded music
sound like the real musical event recorded if the real
musical event is recorded by some plausible and identified
microphone technique.
2 A good speaker is one that in a plausibly obtainable placement
and room condition makes a lot of commercial recordings
sound good to me , good in the sense that I enjoy and am undisturbed
by the sound .
3 A good speaker (or speaker/room combination)
is one that is good in the sense of item 2 but for most
people, not just for me personally.
4 A good speaker is one that makes commercial recordings sound
the way they are supposed to.
5 A good speaker is one that looks good with some
set of usual measurements, whichever set I have decided
or someone has decided are relevant and important.
6 A good speaker is one that a large number of audio people
can be convinced is good.
7 A good speaker is one that operates correctly according
to some theoretical model of what a speaker ought to do.

The problem with 1., as I see it, is that, even with acoustic music, very few recordings are made this way. Most such recordings are either multi-miked or at least made like Mercury and Telarc, with three widely separated omnidirectional microphones. With such miking arrangements, since humans only have two ears separated by about six inches, we have now way of knowing how recordings made with more than two microphones, much less widely separated more-than-two microphones, ought to sound like in terms of the space captured or even the tonality. This means that only recordings made with a single pair of quasi-coincident microphones (e.g., Blumlein, M-S, X-Y, ORTF arrays) should be used to determine the "goodness" of the speakers. My wild estimate is that this limits you to well under a tenth of one percent of all commercially available recordings of acoustic music.

Now you may argue that if you can get those few recordings to sound close to the live event, all other recordings will also sound as they "should" sound. This assumption, while it may be reasonable, has problems. One is that the stereo playback paradigm for the quasi-coincidentally miked recordings must involve a subtended stereo separation angle of at least 90 degrees. From experience I know that with this much separation other types of recordings--the vast majority of them--sound too "stretched out" and frequently have unnatural sounding "holes" in the staging. For most recordings, the engineers assume subtended separation to be around 60 degrees and such reduced separation is needed to avoid such spatial problems.

If possible, a "good" speaker, for the home listener--as opposed to an equipment reviewer-- should be one that in a plausible and obtainable room situation--placement and room acoustics--makes as much commercially recorded music sound like the real musical event as possible. At least for those who enjoy and are not disturbed by the sound of a real musical event, that would make the "good" home speaker a source of maximum enjoyment of recorded acoustic music. I'd argue for combining goals 1, 2, and 3, in other words.

Whether this goal is attainable may depend on how much weight you want to put on strict accuracy to what was recorded versus musical enjoyment of much of the recorded repertoire. It may be a form of the old accuracy vs. musicality debate. I, for one, am willing to give up a bit of accuracy as to the reproduction of the 0.1% of commercial recordings made with a pair of quasi-coincident microphones if that's necessary in order to gain more enjoyment from the other 99.9% of commercial recordings of acoustic music.

One rather obvious concession would be to design the default frequency response of the speakers so as to roll off the highs and boost the bass a bit since it is rather obvious that most recordings of acoustic music through most speakers which are adjusted to measure ruler flat at the listening position sound bass light and treble heavy. It's better to allow home listeners to EQ the 0.1% of properly balanced recordings than to require them to EQ the 99.9% which fit this pattern.

Goal 4. is unknowable. We don't and can't know for most recordings what the engineers heard during the final mastering sessions and we likely don't even know the equipment chain used. I'm assuming that "sound the way they are supposed to" means sound the way the mastering engineer wanted it to sound. There also may have been assumptions made that if recordings sound the way they do on the mastering system, it will sound different, but different in an consumer-preferable way, on most home systems and thus "sound the way they are supposed to."

Goal 5. has some promise. I think that the folks who emphasize the spinorama method are using this. There are even websites which rate speakers based on their conformance to someone's idea of an ideal spinorama data set. One could quibble about how much high-frequency roll off there should be at what angle and starting at what frequency. I would add time coherence as evidenced in a step response on the design axis, plus low distortions of various types at all relevant frequencies and SPLs. I suspect that a data set could be agreed upon which would result in goodness which would only be impeded by listening room problems which could be addressed via speaker/listener placement, EQ, and room treatments.

Note: in a small room like mine, "airless" speakers are quite rare. This may not be so in larger rooms. In my small room, the problem is quite the opposite. The proximity of room surfaces tends to produce overly live/echoey/bright/brittle/nasty sound from most recordings with most speakers without absorbtive or at least diffusive room treatment. Dispersion can be fairly narrow and still sound "room filling" in a small room. The D&D 8c speakers have the closest to ideal dispersion I've yet heard for my small room. Even with bare walls, the sound is well focused and well balanced tonally; it's alive and yet not splashy/bright/annoying in any way, even at high SPLs. It's "awesome," actually, and nowhere "dull" or "airless" in the room or even down the hall. Yes, to me they sound yet better with dispersive room treatment and better yet again with absorptive room treatment at the speaker end of the room, but the differences are not nearly so vast as with most speakers. I could easily live with these speakers with bare walls in this small room.

Goal 6 seems too subjective and too disconnected from the live acoustic music paradigm. Most audio people these days really don't know what live acoustic music sounds like unamplified in a good hall and could care less since that's not their music of choice.

Goal 7 might work if most designers agreed on a theory. But there is little agreement about theory at either the recording or playback end. The lack of standard recording or playback paradigms is what's behind all this uncertainty as to what constitutes a "good" speaker. I think it all comes back to Audio's Circle of Confusion. The lack of such standards creates an inherent ambiguity as to goodness since recording quality has to be judged through playback hardware. See Audio Musings by Sean Olive: Audio's Circle of Confusion from which I quote a part here:


Audio’s “Circle of Confusion” is a term . . . that describes the confusion that exists within the audio recording and reproduction chain due to the lack of a standardized, calibrated monitoring environment. Today, the circle of confusion remains the single largest obstacle in advancing the quality of audio recording and reproduction.


The circle of confusion is . . . Music recordings are made with (1) microphones that are selected, processed, and mixed by (2) listening through professional loudspeakers, which are designed by (3) listening to recordings, which are (1) made with microphones that are selected, processed, and mixed by (2) listening through professional monitors...... you get the idea. Both the creation of the art (the recording) and its reproduction (the loudspeakers and room) are trapped in an interdependent circular relationship where the quality of one is dependent on the quality of the other. Since the playback chain and room through which recordings are monitored are not standardized, the quality of recordings remains highly variable.
 
Continuing on the topic of what is a "good" speaker: Perhaps the ideal are the BBC-designs since those designs were made by engineers who supposedly had the benefit of direct comparison of live unamplified acoustic instrumental/vocal sounds in the recording studio or other recording venue to the speakers reproducing the recordings of this sound in their control room.

But even in this paradigm the Audio Circle of Confusion is still there. The microphone response (at the time the BBC was doing its tests, microphone response was not flat at all), microphone placement (do we know if they used quasi-coincident mics, or was something like the Decca Tree used?), speaker placement, listener placement in both the hall and the control room, control room construction and control room acoustic treatment are all variables. If they were listening to a recording rather than a live mike feed comparison, add the characteristics of the tape recorder and playback machine.

I'm sure the BBC engineers did the best they could, but going from one room to another the best they could do was to adjust all these variables, only one of which was the speakers, so that the sound they heard in the control room was as close as possible to what they heard in the hall.

Basically the BBC was making a lot of assumptions about all the variables involved and then designing speakers to work subjectively as well as possible with the assumptions they made. In other words, they may have settled on listening position in the hall (near the microphone array, I hope), construction parameters of the control room, speaker and listening positioning in the control room, room treatment, microphone type and placement, recording equipment, and then, holding all those variables constant, worked on speaker design to get a closer match between what they heard in the studio and the hall.

This is not a bad method--in practice it's about as good as one can do for designing speakers to reproduce "the absolute sound"--but it's still one fraught with assumptions which can disguise the actual "goodness" of the speakers being designed.

Perhaps it would help me to know more about the BBC's listening comparison process to see how they dealt with all the variables. Did they apply equalization to microphones? What mike array was used? What listening position in the hall did they use for reference? What tape machine and was its response carefully equalized? Etc.
 
I think we need to face the fact that, however good the design paradigm used by the BBC was at the time their work was being done (1970s and before), it is a paradigm no longer available to speaker designers. I don't think any government today is involved in speaker design research. I don't think any speaker designer has daily access to live performances which can be listened to, taped, and compared at will to recordings or live feeds of what is happening live in the next room.

It's worse than that. Very few of those involved in the BBC's original work are still alive, much less still designing speakers. Derek Hughes is one; there probably are a very few others. But today, whatever personal memories these designers have of the process are decades old and accurate auditory memory doesn't last more than a few minutes or hours. Any new designs of these BBC veterans will not have had the direct benefit of BBC-type live vs. recorded comparisons.

What products we have which originated from that age are now quite old and may no longer be performing up to specification. To restore them to specification is possible, I'm sure, but how many restorers know how to do that? And are those products actually the best they could ever be, or are they just the best which could be developed at the time given the theories, designer skill and intelligence, and materials and methods available at the time?

We know, for example, that DSP for crossovers and equalization was not available. If it had been, would closer approximations to "the absolute sound" have been found, or not? Many BBC efforts show a flare in response around the upper crossover point at least off axis; do BBC speakers sound better because of this, or do they sound good despite this if used in an appropriate room with acoustic treatment on the side walls to absorb the non-smooth-response of the off-axis energy?

Perhaps from the BBC research we can still glean design principles which could serve us well in future designs. One of these is probably the thin-wall cabinet design so that cabinet resonances occur low down in the frequency spectrum where they are less audibly intrusive for most folks. I'm sure there are others. But some aspects of BBC designs, such as the double tweeter of the LS3/6 and others with a "supertweeter" kicking in around 13 kHz need to be questioned. Perhaps superior materials and construction techniques available for tweeters today obviate any such design.

My main point here is that the technique used by the BBC to design speakers is no longer available and probably will never again be available. We need to come up with a substitute best design method for approximating "the absolute sound" or at least making speakers with "good" sound. For example, can we identify present speaker designers who are themselves accomplished amateur or professional recording engineers dealing with acoustic music? Such people, it seems to me, would have a clue about whether a speaker they design really sounds something like a live unamplified musical event.
 
I've been reading Toole's book "Sound Reproduction" (3rd edition) lately. Spinorama measurements of speakers and the discussion of those figure prominently in that book. For those interested in seeing such measurements, here is one "mother lode" of such measurements.Click on the link at the bottom of each speaker's "box" to see the spin data.

A collection of loudspeakers measurements (pierreaubert.github.io)
 
I think most of the research and anecdotal experience agrees that smoothness of frequency response is the most important factor in determining whether people find a given speaker to sound "good." Toole finds this smoothness important not only in the on-axis response, but also in the entire "room sound."

I'm not sure how sensitive most people are to percentages of measurable distortion in speakers. Certainly in the low bass high measured distortion doesn't seem that objectionable although some prefer sealed box bass perhaps for its lower distortion. Not all sealed-box bass speakers have low distortion numbers on the order of the old Acoustic Research acoustic suspension types, however. And further up the spectrum?

Some find coherence to be necessary for a "good" speaker. If by "coherence" you mean the inability to hear out various drivers, yes that seems important in small rooms where near-field listening is necessary, but in larger rooms where you can sit back ten feet or more from the speakers it may not be so important.

Some talk about "amplitude linearity" as being a necessary prerequisite for a "good" speaker? Again I'm not sure what is meant by this, but to the extent that it just implies whether the speaker responds uniformly to increased power over its frequency range, I have not seem many speakers deviate from linearity in the SoundStage published anechoic measurements until the SPL gets up to 90 dB or higher, and even then most speakers seem to do pretty well at this.

While I think I can hear phase linearity, the research cited by Toole seems to show that on music signals in ordinary rooms most people either can't hear a difference or if they can hear it, they don't have a clear better/worse preference.

I think radiation patterns are extremely important in real-world listening rooms, as are the rooms themselves. Below the transition frequency, the room is in control and you need to control room modes through speaker and listener placement and/or equalization. Above the transition frequency you need either speakers which have very smooth off-axis response or you need to control what you hear from reflections at the listening position through absorptive and/or diffusive room treatment. In small rooms like mine, I find that I need both smooth radiation patterns and room treatment for "best" sound.
 
When you correct a frequency response of a speaker with DSP you are only or at least primarily applying a correction to what the speaker is doing on the measurement axis. In other words, you can't simultaneously correct the sound the speaker is putting out at other angles.

That is one big reason to look at the shape of the frequency response curves averaged for other angles and all around the speaker--the sound power radiated into the room. If the off-axis response after DSP is still not smooth then what you would hear--especially in a smallish reflective room--could still sound colored because while the on-axis has been DSPed to smooth, the rest of the sound output has not and may actually be less smooth after DSPing the on-axis response to be smooth. And this is one reason why absorptive or dispersive room treatment is necessary for a lot/most speakers in smallish rooms since for a lot of/most speakers the off axis sound is not nearly as smooth as that produced on axis.

Now if all the curves plotted from a spin test seem smooth, the reflected sound will probably not sound so obnoxious and the need for room treatment will be reduced for the speakers to sound good in a typical untreated room. And the room sound will probably remain smooth after further smoothing of the on axis sound with DSP.

Some speakers are said to "ignore the room" they are in more than others--Gradients are one example. Surely part of what makes Gradients and some others easier to live with without purposeful room treatment, even in small rooms, is controlled (narrower) dispersion. However, another part is the smoothness of the response of the off-axis sound. See, for example the off-axis response measured for the Gradient Revolution by Stereophile in Figure 4 at Gradient Revolution loudspeaker Measurements | Stereophile.com The off axis response seems very smooth (although of course downsloping) out to at least 7 kHz. If you compare this graph to most other Stereophile measurements of speaker off-axis performance you will see how special the Gradient Revolutions are in this respect.

The graphs which result from spin tests can give you an idea of the smoothness of the off-axis response of a speaker. Here again is a website which shows the spin test results from many speakers:


A collection of loudspeakers measurements (pierreaubert.github.io)

The spinorama results for the D&D 8c are given in the NextWeb review of the speaker here:

Dutch & Dutch 8c Review: Super high-end speakers with the best bass I’ve ever heard (thenextweb.com)

Now, Toole's book seems to say that room reflections are preferred by trained listeners in his tests, at least for speakers with smooth off-axis response. He also says, however, that this seems to be an individual preference. Some sound engineers, for example, prefer to do their work in rooms with lots of absorption on the room surfaces while prefering to listen casually at home in environments without a lot of added absorption. Equipment or program evaluation may work best in situations which suppress reflections, that is, while listening for enjoyment of the music may benefit from early reflections. I guess I'm always evaluating the sound of the equipment, the sound of the music, or both.:)

However "good" the room reflections from a smooth-off-axis speaker like the Gradient Revolution or 1.4 or the Dutch & Dutch 8c may sound, I still prefer to eliminate the "second venue" constant signature of my small listening room superimposed on whatever recorded acoustics (real or artificial) are on the recording.

So below the transition area I equalize the bass to be smoother but weighty enough to please me on most recordings. For mid and high frequencies I'm relying on acoustic foam near the speakers and diffusion further away from the speakers. I find the "clap track" test on the Sheffield/XLO Test & Burn In CD (out of print--also available on the XLO Reference Recordings Test & Burn-In CD) to be extremely helpful in making sure I have enough absorption in the right places.

Toole's book says that hand clapping has a spectral maximum of 2 kHz or higher and that it takes only about a 1-inch thick foam to damp this adequately. However, in my very small room, played back at healthy levels of 85 dB or so, to really get this track to sound through the speakers as heard from the listening position like it does through headphones (a single transient clap with no post echo at all) takes quite a bit of my 4" thick foam strategically placed.
 
There's a new review of the D&D 8c speakers from Erin's Audio Corner posted on Audio Science Review with quite a few pages of comments as well. See:

 
I've been experimenting a lot lately with different amounts and configurations of my 4-inch-thick acoustic foam. Over time I began to think that perhaps the room was a little too "dead" sounding although the imaging and staging was quite remarkable. I was using a total of 19 2' x 4' batts of the foam, or 152 square feet of it in my 11' x 13.5' x 8.5' room. I had five batts deployed behind and between the speakers, six batts on each of the side walls, and two batts on the floor, one in front of each speaker.

During these experiments I changed nothing about the system or its set up other than the amount and placement of the acoustic foam. I removed all the foam to get a baseline on the sound of the speakers in a reflective room. Not bad at all. As I've said before, the D&D 8c's sound much better in this small reflective room without much room treatment (I'm still using my P. I. Audio Group diffusers on the wall behind the listening seat). Nothing obnoxious at all and a very "live" sound. Imaging and staging are quite nice. I could imagine that many listeners would find nothing at all amiss with the sans-foam sound and might even prefer the "liveness" of the sound.

With this sort of "live" room, the volume does not have to be turned up so high to get the feeling of adequate SPL. The bass is very solid and punchy and the room "lights up" with sound.

However, on classical music with lots of complexity (e.g., the finale to Beethoven's 9th Symphony on RR Fresh) I could hear a bit of compromise in stage depth and image placement/stability. This was probably subtle enough that most people would not notice if they were not familiar with the sound of such pieces with the acoustic foam in place.

Also, there is a subtle "crunch" or "cringe" factor in large scale dynamics at high SPLs. The emphasis is on "subtle" here, though. Such effects have been much more obvious and obnoxious in this room with other speakers when acoustic foam is lacking. I speculate that the combination of the D&D's rather controlled dispersion and the evenness of the off axis response reduces the audibility of such effects and greatly reduces their obnoxiousness.

Again, with the D&D speakers, those not familiar with the acoustic effects of acoustic foam may well not realize that this "crunch"/"cringe" is even there until it is gone. But this is something you never hear from live unamplified music in a decent indoor concert venue no matter how complex or loud things get. Live, the music just gets louder and large scale dynamic shifts are just thrilling with no "cringe" factor at all.

I then began to add back the acoustic foam batts listening for changes with each iteration, also experimenting with their positioning. My goal was to eliminate any "crunch"/"cringe" while improving the "liveness" of the room. To make a long story short, I have now come upon what seems to be an ideal solution. This involved removing the four batts of foam on the side walls nearest the room corners behind the speaker. This removed 32 square feet of foam from the room, leaving 120 square feet.

In the course of this experimentation, I also discovered that, as with the diffusers I formerly used near the speakers, at least in this room the sound of the D&D speakers is sensitive to placement of the foam too close to the vents on the left and right sides of the speakers. A few inches of space is all that's needed. This few inches probably is needed because of the cardioid design of the 100 - 1200 Hz range reproduced by the mid-woofer on the front panel. The sound from this mid-woofer needs to wrap around the cabinet to produce the cancellation of sound radiated to the rear and maintain the intended polar pattern of the speaker. If the foam is positioned too close, the frequency response becomes subtly colored, depth of imaging and expansiveness of the stage in all directions is impaired, and the low end loses some of its magic combination of punch, definition, weight, and extension.

In addition to a bit more "live" feel to the room, the sound now is even better than it was with the additional foam. The bass is even more rightly spectacular, the spatial presentation is even larger, depth of field and width of stage are expanded, and there is a greater sense of envelopment. It continues to amaze me that such sound can be produced from two smallish-looking (from the listening position) speakers!

The new look of the room is similar to the prior one, but here's a few shots of the new foam arrangement.

IMG_7938.jpg

IMG_7939.jpg


IMG_7940.jpg
 
Last edited:
Hi Tom. I really enjoy following your journey with the 8c's. With respect to your latest post, have you considered putting damping panels on the ceiling? Removing early reflections can be beneficial in some ways, but the more you take away, the bigger the effect of the few that remain. I think it would be really useful to show some frequency response measurements at the Main Listening Position (MLP) without and with the damping panels. Really easy to do with the 8c's REW integration. Please do keep the good stuff coming!
 
Thanks for the kind word, Martijn. It's not like I haven't thought about putting foam pads on the ceiling. I've thought about it a lot, in fact.

In my prior two homes' listening rooms I used Sonex foam on the ceiling all the time. Over the years I came to recognize that a little goes a long way up there. Generally, one two-foot-by-two-foot square of it for each speaker's reflection point was plenty. Any more than that started to make the room very dead sounding. I had drywall ceilings and used an electric staple gun to attach the classic Sonex foam to ceiling and walls with staples. Even the 4-inch thick variety could easily be attached this way since the staples could easily be fired through the thin spots in the wedge pattern. Staple holes in drywall were easy to repair with a coat of flat finish paint of the classic Glidden Bone off-white color.

I also used foam near the ceiling in my current room with some prior speakers. There is a centered ceiling fan in this room. I supported the two foam pieces atop the fan blades. That worked okay since I never use the ceiling fan in this room and the foam pads so positioned were, as far as I could tell, just about in the right position given my listening seat and speaker positions. See the picture at this link.

But then I came to realize that the orientation of the fan blades affected the sound regardless of whether there was foam sitting atop the blades. I then took the blades off the fan and that has sounded better to me than having foam atop the blades.

Why not attach foam directly to the ceiling? Well, unlike any other room in the house (it's an old house) this room has a popcorn ceiling finish. That makes it difficult to determine with my reflective mirror method, exactly where to put the foam. I can conquer that problem well enough by putting a level against the mirror to make sure the mirror is horizontal. But that leaves the problem of attaching anything with a mastic backing to the popcorn finish since the ceiling is so uneven.

This is the old part of the house so the walls and ceiling are all plaster and lathe. That makes use of staples impossible to hold up foam and even nails are difficult to drive. The melamine foam I'm now using is much lighter in weight than the Sonex I was using attached directly to the ceiling in prior installations, but it is so delicate that any attempt to press a staple gun into it would break it apart. The right kind of nails could get into the plaster, but the act of hammering them in could also destroy the melamine foam. It seem like mastic is a better possibility

Also, the melamine foam is amazingly difficult to get anything to stick to, even for a few hours or days. Even long strips of duct tape will not stick for long to the flat foam surface. If I wrap strips of tape all the way around the foam, that would create a surface I could attach velcro strips to, but it would create some reflective areas on the foam. But one long wrap of duct tape would probably be enough so it may be worth a try. Some folks online suggest super glue will work to attach velcro strips to melamine foam; I might experiment with that method.

Then, even if I can get the velcro strips to attach to both the melamine and the popcorn ceiling, if I ever decided to move or remove the foam, pulling the velcro off the ceiling would undoubtedly damage the ceiling finish. I've had experience with removing velcro from ceilings and walls before and the result is not pretty, requiring considerable spackle and repainting to repair and the popcorn look would be difficult to restore in the damaged area.
 
  • Like
Reactions: Mensink
Some comments in the latest online review of the Dutch & Dutch 8c speakers bear emphasizing. I don't think I've commented much on mechanical noises from speakers before, but the print and video reviews at Erin's Audio Corner make the point that the 8c's, unlike most other speakers, will play loud (at least as loud as I would ever want to listen, even momentarily) on any music without producing any mechanical noises themselves. This is very true and worthy of additional praise. Few other speakers have been able to do this, in my experience.

Of course, I have not owned any huge horn-loaded speakers which might be expected to do this. But still, even at "sane" levels of 95 dB or below, many speakers I've owned have produced unwanted and obnoxious mechanical noises of various kinds. Many of these noises are momentary and unexpected, occurring just on peaks and thus spoiling the dynamic extremes of which the speaker might otherwise be capable of compassing. With such speakers I find myself in "anticipation mode" wondering when it will occur and at least partially consciously steeling myself against the moment it happens. This ruins my ability to concentrate on the music and enjoyment of it.

Over the years I've owned speakers which:

-"bang" or "pop" on low frequency excursions which exceed their capability (Sanders 10C and Linkwitz Orion)
-buzz from loose or loosening drivers (Linkwitz Orion, B&W Series II Matrix, Ohm Walsh 5, Gradient 1.5 Helsinki)
-buzz from ill-fitting grills (Gradient 1.4) or other undiagnosed problems (Gradient 1.4)
-sound fuzzy or distorted on quiet piano passages and other music from blown tweeters or ribbons (Thiel CS-3, Carver Amazing Platinum Mk IV [several blown ribbons before I gave up])
-hum from transformer buzz (Legacy Audio LFXtreme subwoofer)
-produce flapping noise from woofers pushed past their Xmax (too many speakers to mention)

Probably the worst "noise makers" were the Linkwitz Orions and the Gradient 1.4s. The Linkwitz Orion needed careful SPL control to avoid woofer bottoming and the resulting banging, popping, and flapping. The baffle which held the midrange and tweeter literally flapped in the air back and forth front to back at least a quarter inch, with the result that the tweeter mounting screws were constantly loosening and causing buzzing, requiring retightening. The outboard electronics of the speaker also mechanically hummed and buzzed in an annoying manner.

The Gradient 1.4s, while exceptionally fine sounding when all was well, buzzed like crazy when the magnetically attached metal grills weren't seated just right. Even without the grills mounted, buzzing would crop up occasionally which I could not fix. It was either caused by wiring inside the sphere vibrating against the enclosure, or the sphere itself jumping around a bit against the three rubber bumpers which were supposed to hold it in place (resulting in the orientation of the sphere changing with respect to the lower section over time), or the cable connecting the sphere to the lower section buzzing against the lower section. I could never figure out the exact problem.

The woofers of many speakers I've owned have made flapping noises when pushed too hard at very low frequencies. This seems to be more of a problem with vented speakers than with sealed enclosures. The vintage AR speakers of acoustic suspension design never do this, for example. The AR 303a was astonishing in its ability to produce loud extremely low bass from a relatively small "bookshelf" cabinet.

But the D&D 8c's produce no such obnoxious noises, ever, no matter how hard I've pushed them. Okay, there is a very low level of constant hiss from the tweeter (residual noise of all the onboard electronics) which you can hear from a foot or two from the speaker, but this is constant and not audible even at extremely low listening levels from my listening position. And despite the unusually high levels of bass distortion shown in the NRC measurements for Soundstage, this distortion is not audible to me as such and there certainly is never any spurious mechanical flapping noise from the sealed-box twin 8-inch subwoofers. It is extremely refreshing to be able to relax and trust the speakers to be able to handle whatever I throw at them with complete aplomb.
 
Last edited:
I think I have finally figured out a way to put absorbing foam on my popcorn ceiling without doing damage to it. It seems to be holding; we'll see if it stays up.

It is a definite sonic improvement when combined with removing the foam that was centered on the wall between and behind the two speakers. Basically I put two 2' x 2' sections of foam on the ceiling in the first reflection areas and removed the 2' x 4' batt of foam I had at the center of the wall behind the speakers which was not really in any first reflection area. This change significantly improves the three-dimensionality of the presentation and further increases its apparent size while leaving the liveness of the room at about the same level as it was before since the total amount of foam placed against the room surfaces is the same as before.

First I determined the first-reflection areas on the ceiling by trial and error using my 4" x 6" flat mirror mounted flat agains the ceiling temporarily with masking tape. Once I could see the front of each speaker in the mirror when I was sitting in the listening position I used blue painter's tape to mark the outline of the mirror. Then, after removing the mirror, I put a small piece of blue painter's tape in the center of the mirror outline. I then measured one foot in four directions from that centered piece of tape and put additional pieces of tape on the ceiling at those points. That gave me the outline of the area where I needed to mount a 2' x 2' piece of foam on the ceiling.

I then cut one of my 2' x 4' batts of foam into two 2' x 2' pieces with my Cuisinart electric kitchen knife. The foam cut like butter with that knife. Then I wrapped a continuous strip of two-inch-wide white duct tape around the center of each white piece of foam so that the tape overlapped a bit after encircling the foam and could thus attach to itself for a few inches. That will prevent the tape from coming unstuck from the foam.

Then I mounted two 4" x 2" pieces of heavy duty outdoor velcro meant for rough stone or concrete surfaces to the ceiling and made sure that it was stuck as tightly as I could get it to the popcorn ceiling. I left the protective coating on the velcro which would attach to the foam as I pressed the velcro into the ceiling. Then I removed the protective coating. I then carefully placed the foam against the sticky side of the velcro so that the duct tape wrap made contact with the sticky velcro. The duct tape, unlike the foam, should stick well to the velcro. That's the plan at least. I have high hopes that it will stay on the ceiling and not fall down.

P.S.: They did not fall down, but I did add additional support by attaching an additional 2" x 4" piece of this same velcro to the ceiling just touching the outside edges of the foam where the tape wraps around. I then anchored additional duct tape to the exposed sticky side of the mated velcro pads and ran this tape along the existing tape going around the foam batt. The foam is thus now held to the ceiling both by velcro attached to the ceiling under the foam and velcro attached to the ceiling just outside the edges of the foam. The foam is not going anywhere now.
 
Last edited:
There are many who think that speakers which supposedly have "ideal" measurements in spinorama tests in fact have off-axis dispersion which falls off too quickly in the high frequencies to sound openly natural in your listening room. The thought is that such speakers will sound "dead."

Comparing the spin test results of the number-one-rated KEF Ref 5 and D&D 8c at https://pierreaubert.github.io/spinorama/ you can see that for the 8c the on-axis and listening window (defined as plus or minus 30 degrees horizontal, plus or minus 10 degrees vertical) are practically identical. For the KEF, the listening window is a couple of dB below the on-axis in the treble.

And then the early reflections of the D&D basically don't drop below -5 dB from the on axis and the sound power is about -8 dB all the way to the top of the band. For the KEF, early reflections continue to drop off to more than -7 dB and sound power to below -10 dB.

Clearly the D&Ds don't drop off in high frequencies as much as the KEF. Note that this is confirmed by the fact that the estimated in-room response of the D&D only slopes down about 7 dB from 20 Hz to 20 kHz, while that of the KEF slopes down at least 10 dB.

This is all subjective and perhaps room size and set up dependent. Some have said they don't think a room the size of mine could ever sound very good since the speakers have to be so close to the side walls. That may be correct in the sense that I think the greater amount of horizontal dispersion in the lower highs favored by some throws too much high frequency energy toward the sidewalls for a room my size. The side wall reflections might be too strong to be satisfactorily dealt with by the amount of absorption one can easily apply to mitigate the problem.

But remember that I listen exactly on axis and stay put. What I hear from the listening seat is strongly governed by the on-axis or at least listening window response. And I've also found that in my room the D&D speakers sound better than any others I've had in this room when all the room treatment is removed. I attribute that to the combination of its dispersion pattern and the fact that the off-axis response is so very smooth all the way out to the highest treble, unlike most other speakers--especially BBC-legacy speakers like the Harbeths which are anything but smooth in off-axis response in the presence range.

And I'll also say that, as
Mitch Barnett said in his review, the D&D 8c really is unique in my experience for having a "just right" treble balance for a lot of material. Yes, this speaker is a bit more high-frequency subdued than many/most others, but that is a plus to my ears and allows comfortable high-SPL listening to a lot of material which gets a bit nasty/grating on other speakers--such as high brass and big band recordings. Massed high strings also benefit and lack the wirey edge and what sounds like high frequency distortion floating above the string section. Recordings sound like real string sections in real life with real air and projection, not distortion floating above. Part of that may be the time alignment, but I think another part is just proper high treble level in both the direct and off-axis sound.

Yes, the Harbeths and Gradients are also among the best speakers I've heard, as I've noted. It's just that I find the D&D 8c's to be better yet in all ways, including overall balance and high frequency balance. I am almost never bothered by excess high frequencies at live unamplified concerts, no matter the venue, even though different halls and different seats in the same hall obviously have much different frequency balance. To my mind, that's the way most recordings should behave through your speakers at home as well. Yes, you can immediately hear the changes in tonal balance from one track to another, but mostly nothing sounds obnoxiously bright or bass thin on the D&Ds.
 
Here are a couple of wide-angle pictures of my current D&D 8c set up showing the deployment of four-inch-thick Alphasorb absorbing acoustic foam at the first-reflection areas on the floor, ceiling, wall behind the speakers, and side walls, as well as the diffusive treatment of the listening end of the room.

IMG_7989.jpg

IMG_7990.jpg
 
Last edited:
  • Like
Reactions: Addicted to hifi
If you look at the spin test results of the KEF Ref 5 and D&D 8c at https://pierreaubert.github.io/spinorama/ you can see that for the 8c the on-axis and listening window (defined as plus or minus 30 degrees horizontal, plus or minus 10 degrees vertical) are practically identical. For the KEF, the listening window is a couple of dB below the on-axis in the treble.

And then the early reflections of the D&D basically don't drop below -5 dB from the on axis and the sound power is about -8 dB all the way to the top of the band. For the KEF, early reflections continue to drop off to more than -7 dB and sound power to below -10 dB.

Clearly the D&Ds don't drop off in high frequencies as much as the KEF. Note that this is confirmed by the fact that the estimated in-room response of the D&D only slopes down about 7 dB from 20 Hz to 20 kHz, while that of the KEF slopes down at least 10 dB.

This is all subjective and perhaps room size and set up dependent. Some think a room the size of mine could ever sound very good since the speakers have to be so close to the side walls. They may be correct in the sense that I think the type of horizontal dispersion favored by many throws too much high frequency energy toward the sidewalls for a room my size. The side wall reflections might be too strong to be satisfactorily dealt with by the amount of absorption one can easily throw at the problem.

Also remember that I listen exactly on axis and stay put. What I hear from the listening seat is strongly governed by the on-axis or at least listening window response. And I've also found that in my room the D&D speakers sound better than any others I've had in this room when all the room treatment is removed. I attribute that to the combination of its dispersion pattern and the fact that the off-axis response is so very smooth all the way out to the highest treble, unlike most other speakers--especially BBC-legacy speakers like the Harbeths which are anything but smooth in off-axis response in the presence range.

And I'll also say that, as Mitch said in his review, the D&D 8c really is unique in my experience for having a "just right" treble balance for a lot of material. Yes, this speaker is a bit more high-frequency subdued than many/most others, but that is a plus to my ears and allows comfortable high-SPL listening to a lot of material which gets a bit nasty/grating on other speakers--such as high brass and big band recordings. Massed high strings also benefit and lack the wirey edge and what sounds like high frequency distortion floating above the string section. Recordings sound like real string sections in real life with real air and projection, not distortion floating above. Part of that may be the time alignment, but I think another part is just proper high treble level in both the direct and off-axis sound.

Yes, the Harbeths and Gradients are also among the best as I have noted. It's just that I find the D&D 8c's to be better yet in all ways, including overall balance and high frequency balance. I am almost never bothered by excess high frequencies at live unamplified concerts, no matter the venue, even though different halls and different seats in the same hall obviously have much different frequency balance. To my mind, that's the way most recordings should behave through your speakers at home as well. Yes, you can immediately hear the changes in tonal balance from one track to another, but mostly nothing sounds obnoxiously bright or bass thin on the D&Ds.
 
If attacks are so important (and I'm not saying they are not), wouldn't time coherence of the direct response of speakers at the listening position be the most important thing? Yet research shows that this apparently is not very audible to most people.

I happen to agree that speakers which produce a coherent-looking impulse or step response have that extra "something" which makes the sound more coherently natural. But outside of a few manufacturers over the years and some recently DSPed models like the Dutch & Dutch, few acclaimed speakers have been time coherent in this way. Certainly no Infinity, Revel, JBL, Genesis, Magnepans, Wilson, Linkwitz, or Sonus Fabers are. The audiophile-oriented ones that measured as coherent in this way before DSPed speakers included Quad 63 and its derivatives, Dunlavy, Spica, Vandersteen, Thiel, Meadowlark, Sequerra, and maybe a precious few others I'm forgetting, but not many.

Obviously no BBC-inspired speakers are time coherent in this sense. The B&W DM6 (if you count that as a BBC-inspired speaker, which is open to serious question) is an exception.

Just because speakers have drivers physically offset to appear to align their voice coils does not mean that they produce a time coherent response at the listening position. For multi-driver systems driven through an analog crossover you also need a first-order (6 dB/octave) crossover. The Quad 63 is a special case with an analog domain delay line which accomplishes the same thing. While the B&W DM6 model was apparently time coherent, the later 801, 802, etc. had measurements and specs which clearly indicate they were not time coherent.

Also, I think it would be very difficult to time align any subwoofered system which has more than one sub per channel, especially if the subs are at unequal distances from the listener.

Of course, other factors besides time coherence may be subjectively significant to realism of attacks and decays, such as lack of box resonances, lack of listening room reflections, and the signal-to-noise-and-distortion ratio of the listening environment/electronics/speakers (especially for decays).
 
In my experience, panel speakers and full-range dipole speakers in general are sensitive to the in- and out-of-phase reflections off wall surfaces. Linkwitz used to recommend Rule of Thirds positioning for dipoles for just this reason. Such positioning means that if the dipole is not toed inward the reflection off the wall behind the speakers will always be in phase at the listening position with the reflection off the wall behind the listener. You can use absorption and/or diffusion to break up these reflections.

The problem is, especially with dipole panels, you have to toe them in significantly toward the listening position to hear the highs because of narrow horizontal directivity. This brings the side wall reflections into play and changes the path length between reflections off the wall behind the speakers relative to the path to the wall behind the listener. Now more absorption/diffusion is necessary and you must deaden the room considerably to reduce the "echo" effect. If your room is big enough, this "echo" can be perceived as spaciousness enhancing. If it is not large, it's just obnoxious echo overlaid on all program material. Thus my abandonment of full-range dipole speakers for my small listening rooms.
 
Let's start with the assumption that a speaker's frequency response is its most sonically defining characteristic. I think it is then astonishing that some reviewers seem to be saying over the last few years that we should throw that out as a criteria for determining what to buy since we can EQ the speakers anyway.

Now this wouldn't seem so odd to me if experts could tell us with a reasonable degree of certainty from looking at a reasonably complete set of frequency response measurements whether a particular speaker's evident frequency response problems could be EQed away by home users. But that does not appear to be the case--at least not yet.

So it still seems uncertain how much difference EQ can make even when we have reasonably good measurements to look at. That is not a strong argument for abandoning the traditional method of buying speakers which is to listen to them without EQ either at a dealer or at home. Usually, in either situation, one doesn't really have time to determine what needs fixing and then to evaluate the attempt to fix it with EQ.

Thus we are at best back to the conclusion that one should not buy any new speaker unless some reviewer you respect has tried it for an extended period in his home and has tried EQ to fix any flaws and if so with what ultimate result relative to other comparably priced speakers.

Here's another uncertainty about the application of EQ. The off-axis response and on-axis response can't be simultaneously adjusted. Some folks seem to prefer off-axis curves that looks like what some call the "dying swan," where the first reflections and sound power begin rolling off far below 2 to 4 kHz. Some reviewers seem to like speakers whose far off axis response holds up until 2 to 4 kHz (kind of like a concert hall) and then rolls off. I have suggested that this might well be room dependent; in my small room I seem to prefer speakers with narrower HF dispersion. This could be much different (even for my tastes) in a larger room where the speakers can be kept at least five feet from the side walls.

I don't think anyone is arguing that the on-axis response should be anything but flat from close up. Maybe even the listening window response (defined these days as plus or minus 30 degrees horizontally and plus or minus 10 degrees vertically) should be flat. The D&D 8c speakers have very flat on-axis and listening window response. It is only further off axis where they may take more of the "dying swan" approach.

If one always listens from the sweet spot, and one always aims the speakers at one's ears in the sweet spot (as I do), what difference does the off axis response make if you are listening from fairly close up (say 5 to 7 feet)? Well, from experience in my small room, to my ears it does matter in terms of the amount of effect room treatment has on the sound.

And I know that if the speaker gets REALLY directional (e.g., the Janszen Valentina Active I had in this room with the side-firing Air Layer tweeters turned off), even in this small room (or maybe especially in this small room) I do not really like the result since I'm disturbed by the sound seeming so muffled (dying swan-like) when I enter and exit the room and indeed from most any spot a foot or so away from the sweet spot. No, even in this small room, I prefer coping with wall reflections than overly dead sound outside the sweet spot. But there seems to be a happy medium and speakers such as the Harbeth M40.2, Gradient 1.4, and D&D 8c are all within range of that happy medium, with my favorite being the bit-more directional D&D 8c.
 

About us

  • What’s Best Forum is THE forum for high end audio, product reviews, advice and sharing experiences on the best of everything else. This is THE place where audiophiles and audio companies discuss vintage, contemporary and new audio products, music servers, music streamers, computer audio, digital-to-analog converters, turntables, phono stages, cartridges, reel-to-reel tape machines, speakers, headphones and tube and solid-state amplification. Founded in 2010 What’s Best Forum invites intelligent and courteous people of all interests and backgrounds to describe and discuss the best of everything. From beginners to life-long hobbyists to industry professionals, we enjoy learning about new things and meeting new people, and participating in spirited debates.

Quick Navigation

User Menu