@rauliruegas wrote:
" Even the AK SWARM manufacturer/owner posted in this forum:
""" You can get good bass in one sweet spot with two equalized subs """. Not 3 or 4 but TWO."
I probably said the first part, but just to be clear I DID NOT say the second part ("Not 3 or 4 but TWO"). Imo three or four subs intelligently positioned are superior to two, whether or not EQ is involved.
Quoting rauliruegas again:
"Again, even the AK designer/owner from where came your 4 bass array configuration posted in this same forum that two subs for two channel stereo MUSIC are ENOUGH."
I don’t remember saying that in so many words, and if I did, the word "ENOUGH" is highly subjective and I should have been careful how I used the word.
The basic problem is, the room imposes a significant peak-and-dip pattern on a subwoofer at low frequencies. The peaks are especially undesirable because they take longer to decay and can make the bass sound slow, boomy, one-note, etc.
We can move the subwoofers and change the peak-and-dip pattern, but we cannot eliminate it by positioning alone.
We can use bass traps to absorb bass energy and reduce the magnitude of the peaks and dips.
We can use EQ to reduce the energy going into the peaks and maybe even boost the energy going into the dips, which can work well at one location, but will usually make the bass worse at other locations because the peaks and dips will be at different frequencies.
We can use multiple subs spread intelligently around the room so that each produces a significantly different room-interaction peak-and-dip pattern, and the sum of these different patterns will be pretty smooth, and this smoothness will hold up well throughout the room. In general two subs intelligently distributed are twice as smooth as one, and four subs are twice as smooth as two.
I see no reason why these different approaches cannot be combined.
I really haven’t spent much time figuring out the best way to use two subs, because imo the improvement in going from two to four is worthwhile (assuming approximately the same total expenditure). If someone is limited to using two subs, then I suggest they go to one of my competitors. Not that I’m an expert on all of the subs out there, but ime Rhythmic makes some very nice ones.
Duke
guy who makes the Swarm |
Tyray wrote: "When your Swarm first came out I and bunch of others were impressed at such a brilliant thinking out of the box idea. I completely believed and understood the science behind it." Thank you very much. If there was any "brilliant" thinking on my part, it was in quickly recognizing the merit of the distributed multi-sub concept when Earl Geddes described it to me, and in immediately asking him if I could license his idea (he replied, "No, you can just use it," so I try to give him credit when the opportunity arises). I also appreciate the Richard Vandersteen quote you posted above: "... multiple subs automatically give a linier response in the room. This should be amazing as we have always been a supporter of multiple subs." Indeed, I recall Richard recommending stereo subs long before I heard about using four distributed subs from Earl. Was Richard the first to propose that two subs are significantly better than one? I think he was. I have always liked Richard’s speakers very much, and feel honored that a few of his customers have chosen my somewhat unorthodox little subwoofer system to go with them. Duke
|
@noble100 wrote: "Duke... I was hoping to get your honest opinion on my thoughts on how I understand multiple sub systems function in general as well as my understanding of how bass is recorded on CDs and vinyl."
In general I agree with what you wrote, so let me just toss out a few comments.
" We all are unable to localize deep bass frequency soundwaves..."
My understanding is that’s generally true in a room. The figure I use is 80 Hz, rather than 100 Hz. I think Floyd Toole uses 80 Hz. This doesn’t necessarily mean that one cannot detect the location of a sub which is crossed over significantly lower, say at 40 Hz, because crossovers are not brick walls, so upper bass/lower midrange energy can give away a sub’s location if it comes through loud enough. Therefore in my opinion a steep lowpass filter on the sub helps to hide its location.
"the bass is summed to mono on frequencies below 100 Hz on all vinyl and cd recordings."
I wouldn’t say "all", but I would say "almost all". (If we’re talking about a Swarm/DEBRA system, a second amp can be added to give you true stereo bass).
"Our brains are able to associate the fundamental deep bass frequencies reproduced by the subs, that are not able to be localized, with the deep bass’s higher harmonic frequencies, that extend well beyond 100 Hz, which are reproduced by the main speakers that are able to be localized. This psychoacoustic association allows us to localize the deep bass in the soundstage, for example the kick drum is located in the rear center and the upright bass is located in the front to the left, which would not be otherwise possible without this psychoacoustic association our brain’s are capable of."
I agree.
Duke |
@noble100 wrote: "Are you making a distinction between the perceived true stereo bass that I'm currently experiencing and your mention of "a second amp can be added to give you true stereo bass"?" IF the recording has true stereo bass, adding the second amp would allow you to reproduce left and right channel bass separately, instead of summed. That's a big IF.
"On several other threads no one, including myself, was able to name a single recording on any format containing true stereo bass. Are you aware of any?"
I don't own any, but my understanding is that Ralph Karsten has said the vinyl recordings he makes have stereo separation all the way down. And I believe him! Duke |
Audiorusty asked: "A question for the DBA folks; How do you get the mids and highs to spread out evenly through the room to match up with the bass at all locations in the room? Also does a DBA affect impact?" The bass doesn’t sound like it’s coming from everywhere. The bass sounds like it’s coming from wherever the bass instruments are in the soundstage. We get our localization cues from the higher frequencies that the main speakers produce. In my experience impact is improved with a good distributed multi-sub system. This is because smooth bass is "fast" bass. We hear peaks as "slow bass" or "boomy bass", and we hear dips as "weak bass" or "lack of impact". (Dips are usually less objectionable than peaks.) Our natural instinct is to think that having multiple arrival times from the various subs degrades the bass impact, but that’s not what happens. The ear DOES NOT EVEN DETECT the presence of bass energy from less than one full wavelength, and bass wavelengths are many feet long. So by the time we even begin to detect the presence of bass energy, the bass has already reflected off of room surfaces multiple times. We do not hear a "first arrival sound" in the low bass region because by the time we hear the bass, we are already hearing the room’s effects. But what happens to the trailing edge of the notes (how quickly and smoothly they decay) is very important to the sense of impact, and this is where a good distributed multisub system excels.
To put it another way, we cannot hear bass apart from the room’s (detrimental) effects, so we might as well take the room into account. By spreading multiple bass sources around the room, one might say that we get the room to work "with us" somewhat instead of "against us".
Last year at the Rocky Mountain Audio Fest we got a nice compliment from an extremely competent cable manufacturer with decades of audio show experience. He asked us to play his reference recording of "Fanfare for the Common Man". He said ours was the most natural reproduction of that recording he had heard at a show, and in particular he said "THAT is what a tympani sounds like - THAT is what I hear at the symphony". Duke |
@rauliruegas wrote:
"Position at a normal one seat position and play music ( digital or analog ) with very low bass recorded information and then through several tests/evaluations lest us know if you can or can’t detect from where comes the sound you are listening... Main speakers off(no sound.)."
Well of course under those conditions (playing music through the subs with the main speakers OFF) you can hear where the subs are! This is because the lowpass filter on the subs are not brick walls.
But with the main speakers on, the output of the mains will be so much louder than the output of the subs that it will effectively mask the location of the subs, even if the subs are letting a little bit of upper bass/lower energy leak through.
The point is not whether there is an unrealistic test condition in which a distrubuted multisub system falls short. The point is, how does it perform when used under normal conditions?
Duke
|
@bdp24 asked:
"Is for the subs to be wired/operated in mono mandatory in a swarm sub system?"
Not at all.
"In other words, does the swarm not "work" (reduce/minimize bass peaks and nulls in the room) if the subs are run stereo?"
Works just fine in stereo. The only reason my commercial Swarm system is normally mono is, I can hit a more attractive price point by driving all four units with a single amplifier. As an option the Swarm can be supplied with a second amplifier.
"Or is it just a matter of that with a steep enough filter (say 4th-order, 24dB/octave), there is nothing to be gained by running the subs stereo?"
If the bass is summed to mono on the recording anyway, which I believe to almost always be the case, then imo there is nothing to be gained by running the subs in stereo. But imo there is something to be gained (greater sense of envelopment) by using introducing a roughly 90 degree phase difference between the subs on the left-hand side of the room and the subs on the right-hand side of the room, and doing so requires two amplifiers, in which case you might as well do stereo. That way if you know you have a recording with true stereo separation all the way down, you can quickly dial back in normal phase using the phase controls on the amps.
"Is running two subs at the front of the room (in the neighborhood of the loudspeakers) in stereo, and the third and fourth in mono, a viable option? THAT would make using a shallower low-pass x/o filter on the subs (for whatever reason) possible."
You can do that, but imo it doesn’t really make a shallower low-pass filter more feasible. Imo the main reason for using a steep low-pass filter is to prevent the subs from passing upper bass and lower midrange energy loud enough to betray their locations.
If your subs have fairly shallow (12 dB/octave) built-in lowpass filters, you might try this: Use a lower low-pass filter frequency for the subs that are closer to the listening area, since these are the ones most likely to be heard as separate sound sources in the upper bass/lower midrange region.
Duke |
@bdp24 wrote: "would you expand on your recommendation of introducing a 90 degree phase difference between left and right subs?" This is an idea I learned from a paper written by David Griesinger many years ago. David recommended two subs, one directly to the left and one directly to the right of the listening area, up against the side walls. With the phase 90 degrees apart (in "phase quadrature"), this would approximate at the left and right ears the somewhat out-of-phase low frequency waves one would experience in a large hall. The result is a greater sense of being immersed in a large acoustic space, which means the ear is not perceiving as much "small room signature" super-imposed on top of the recording. As you hear less of your room’s signature, you correspondingly hear more of the ambient space on the recording. Wouldn’t say it’s a huge difference, but as long as it doesn't bust the budget I think it’s arguably a worthwhile improvement. Duke |
Audiorusty asked: "Since a DBA system can evenly distribute low bass evenly to all parts of a room are you also able to distribute the mids and highs evenly to all parts of the room or are you still relegated to a single listening sweet spot in the room due to the directionality of the mids and highs?" Are you asking about speakers in general, or about my speaker designs in particular? Duke |
Audiorusty wrote: "With as good as a DBA system is, there still will be only one sweet spot in the room." Imaging will of course be best in the sweet spot. But it is possible for the imaging to hold up fairly well across a fairly wide listening area, and for the tonal balance to hold up well just about throughout the room. This all depends on the directional characteristics of the main speakers and how they are set up, and the details are somewhat counter-intuitive. * * * * Raulirugas wrote: "I’m just talking about the myth that we can’t detect the direction for frequencies 80hz down ( two channel with two subs. ) and this is not true." Detectable under specific conditions that are not representative of listening to music reproduced in a home listening room, yes. Worth trading off the advantages of a good distributed multisub system for, well I guess that depends on your priorities. Just about everything in audio involves tradeoffs. Make an informed choice and pick the set of tradeoffs that makes you happy. Fortunately in audio if a particular choice doesn’t make you happy, you can choose again. Duke |
Mijostyn wrote: "The statement that the ear (which should be ears) have very poor time domain response is 180 degrees wrong." You are correct! I did not proofread carefully. Here is what I should have said: "Because the ear has very poor time domain response AT LOW FREQUNCIES..." The ear is indeed quite sensitive in the time domain at higher frequencies. Mijostyn again: "Speakers + Room do not equal a minimum phase system." Yes they do at low frequencies, which is what I said. This according to Floyd Toole and Earl Geddes. The in-room frequency response tracks the time domain response. The phase behaviors of individual reflections don’t matter when viewed in isolation, but their sum is relevant as it shows up simultaneously in both the in-room frequency response and in-room time domain response. It is the sum that we hear. Mijostyn: "With the drivers acting as one the bass drum impact will strike the listener in phase with the greatest force resulting in the largest smile. The decay afterwards is of no great significance." This is what our intuition tells us, and our intuition is wrong. We literally cannot detect the presence of bass energy from less than one wavelength, which is 22.5 feet at 50 Hz (ballpark resonant frequency of a bass drum). A study was done in which less than one full wavelength of low frequency energy was played over headphones, so there were no room reflections, and it was UNDETECTABLE. And it took MANY cycles before the ear began to register the pitch. By the time your ears BEGINS to hear the impact of that bass drum played over your system, so much time has passed that any minor arrival time differences are inconsequential. The decay is of enormous consequence because it shows up as a frequency response peak. And this is because speakers + room = a minimum-phase system at low frequencies. If the decay is slow at some frequency then the bass sounds fat because there is a response peak at that frequency. Mijostyn: "I know of one person who set up his four subs as I suggested and he thought it made an improvement." The improvement may not have been for the reasons you suggest. I strongly suspect that what happened is, the frequency response improved. The in-room frequency response is what dominates our perception at low frequencies (though the in-room frequency response is merely a manifestation of the in-room time domain response, and vice versa). Duke
|
Mijostyn wrote: "For the drivers to act as one they can be no farther apart than 1/2 the wavelength of the crossover point." You do an excellent job of explaining your suggestion, but imo the issue it addresses is not one that a distributed multisub system is concerned with. With a distributed multisub system, "the drivers acting as one" is definitely NOT the goal. We want each sub to interact with the room very differently from the others, and that is accomplished by spreading them far apart. So not only are there subs more than a half wavelength apart, they will also (almost always) have different path lengths to the listener(s). Because the ear has very poor time domain response, such that it cannot even detect the presence of bass energy from less than one wavelength, a bit of smearing in the initial arrival times of the different subs is inconsequential. What DOES matter a great deal from a perceptual standpoint is how the notes decay. Since speakers + room = a minimum-phase system at low frequencies, when we have fixed the frequency response we have simultaneously fixed the time-domain response. Imo this is something that a good distributed multisub system does well. (The ear hears low-frequency ringing quite well, but such ringing is always accompanied by a frequency response peak, and vice-versa. Anywhere in the bass region where there is a peak, whether it originates from a speaker anomaly or a room anomaly, the energy in that region takes longer to decay. It is actually the frequency response peak that we hear, not the ringing - the slow decay - itself. But when we have fixed the one, we have simultaneously fixed the other.) Duke |
Mijostyn wrote: "Duke I need you to define what you mean by a minimum phase system." I probably shouldn’t have used that term, as it sounds like I’m talking about phase response. You can google the term "minimum phase system" if you want. Instead of using the term I should have simply said something like this: At low frequencies, there is a correspondence between the frequency response and the in-room decay such that they track one another. When there is a frequency response peak, the corresponding decay time is longer. The correspondence is such that if we fix one, we simultaneously fix the other. Misostyn again: "Duke is saying is that placing subs randomly throughout the room creates a situation that smooths out the frequency response throughout the room." Just one minor point: I’m not really advocating RANDOM sub placement; I’m recommending DELIBERATELY ASYMMETRICAL and INTELLIGENT sub placement. Mijostyn: "According to Earl Geddes... you want to keep your subs as far apart as possible and throw the fourth one away." The four-sub configuration I use was Earl’s first-generation multisub-system. He subsequently wrote a program that can analyze the outputs of three subs and design custom digital equalization filters that will attain his goals without needing that fourth sub. The drawback is, if you move any of your subs or buy a new sub or change rooms, you need to have Earl redo the equalization in order to return to optimization. Hleeid wrote: "Still looking for comparisons between the Swarm and DEBRA systems." The only significant difference is the shape of the footprint: The Swarm units have a square footprint (12 inches by twelve inches), and the DEBRA has a rectangular footprint of approximately the same total area. The woofer is the same, the internal volume is virtually the same, and the port tuning frequency is the same. They use the same kind of long, pluggable port. In the Swarm, the downfiring port is behind the woofer, and in the DEBRA the downfiring port is alongside the woofer. Hence the different footprints, but I don’t think there is any corresponding difference in performance. Jim thought a rectangular footprint looks better, and he has a much more artistic eye than I do. I sent the Swarm to Robert Greene of The Absolute Sound for his evaluation, which is why the Swarm rather than the DEBRA has been named in three Golden Ear awards and a Product of the Year award. Duke |
" What is the smallest room size recommended for use with the DEBRA/SWARM systems? "
This will be somewhat counter-intuitive, but a small room benefits more from a distributed multisub system than a large room does. The reason is, smaller rooms are correspondingly worse in the modal region and so they have more room for improvement!
So the large room starts out better. Assuming equivalent multisub systems in both, the small room will improve more and "catch up" to the larger room somewhat, but I would expect the larger room to still sound better.
I think the lower room size limit is set by practicality... when you can just barely shoehorn four small subs into the room along with the rest of the system and yourself, you have found the smallest room size.
I’m working in a small-room optimized speaker system and surprise surprise it includes four small subs, two of which are built into the bottoms of the main speakers for the sake of practicality.
Duke |
Hi Tim,
My reasoning is simply this: Four one-foot-square footprints (2 mains w/built-in subs + 2 separate subs) are easier to shoehorn into a really small room than six (two stand-mounted mains + four separate subs). And I already make the latter.
If we use a little bit of asymmetry in the placement of the main speakers those first two subs are now asymmetrical, even if they are tied to the main speakers. And we have other tricks up our sleeve if needed... such as polarity reversals, or adding a second amp and dialing in phase quadrature.
The main driver arrays are fairly directional so they can be placed close to the walls with relatively little downside. Compared to most small speakers, they have less need of distance from the walls in order to get good soundstage depth. So their requirements probably are not going to screw up subwoofer placement all that much.
And with four distributed subs, the specific locations of the individual subs are a lot less critical than with one or two subs. So imo there is less penalty from those first two subs being stuck underneath the main speakers.
I hope to end up with a competitive system for those who can accommodate four footprints in a small room.
Duke
|
"It seems like your new system concept will probably work well not only in a variety of small rooms, but also some medium sized rooms."
The main speaker array has a lot more dynamic range than one is likely to use in a small room, so yeah the system should work well in a medium-sized room too. Virtually everything in speaker design being a juggling of tradeoffs, I did make a few in favor of small room optimization.
"Does it matter if the woofer is layed down horizontally on the floor?"
Nope! I have several customers who use at least one Swarm unit on its side, sometimes on the floor, sometimes atop a book case.
"Leaning towards the DEBRA system since the (flatter?) profile woofers would probably take up less floor space."
Sounds good to me! (For the record, I don’t get a "kickback" on Debra sales.)
"Duke - are you building (or have built) a new sub array system for smaller rooms?"
Not yet, but shoot me a message or an e-mail if you’d like to exchange ramblings about tradeoffs.
Duke
|
"Duke, I think what people are concerned with is the size of the subs in a small room."
That makes sense, but I don’t think I can offer a practical solution.
You see, in part because I’m a small company, my costs don’t scale down proportionate with enclosure size. So as I start shrinking the boxes I start to become less competitive with four small off-the-shelf subs.
Duke
|
"...the market for smallish subs might be higher than you suspect."
I’m not saying I’ll never go there, but it has to be something that I believe offers worthwhile improvement over its competition, rather than just being four small subs sold as a package.
For instance, the most obvious feature of the Swarm is, it’s four small subs sold as a package. But when we get into the performance specifics, imo the individual units are competitive with their ballpark price equivalents.
Duke
|
" My main concern is to what degree would the extremely good bass power, impact and dynamics qualities of the four separate subs configuration used by the SWARM and DEBRA systems would be compromised when two subs are incorporated into the main speakers or smaller bass drivers are utilized. "
James and I have done a couple of audio shows with systems that incorporate two of the four subs into speaker stands, which is conceptually similar. One such room was where an industry veteran cable manufacturer said our room did the best job on his reference Fanfare for the Common Man recording of any room he had heard at any audio show, and in particular he said the tympani sounded like what he hears at the concert hall.
So I won’t claim zero compromise, but I don’t think there’s very much.
Duke
|
"So for those of you enjoying your DBA systems, any suggestions on creative speaker cable management?"
You can use 16 gauge zip cord, make the wires as long as necessary to hug the walls if that helps, and you can run ’em under rugs.
|
Clio09 wrote: "I don't run equal lengths of cables to any of my woofers. Instead I cut
to the length required for each woofer. I also don't worry about
stepping on the cables while they are under the carpet."
Same here.
Duke
|
