Is there such a thing as "box sound"?

[Soniclife]

I have no idea what they mean with that but personally I tend to hear speakers sound 'boxy' if the 100-300Hz region is dominant (= there is a peak there for whatever reason)
Especially around the 100Hz region it tends to 'hurt' my ears
Once correct DSP (EQ) is applied it goes away immediately

I hear something similar, but to a few 100hz higher up, but like you EQ is a total solution.

If the boxy thing is real, and the people making the claim need to prove it, my guess is it's the change in directivity from omni at low frequencies to forward radiating when the baffle takes over. Are cardioid speakers also perceived as boxy, when people cannot see the box.

 

[Flaesh]

unscientific

Islands in the Stream by Hemingway. In the beginning of part 3 sombreros cientificos\scientific hats are mentioned. May be this may help.

 

[3ll3d00d]

I hear something similar, but to a few 100hz higher up, but like you EQ is a total solution.

If the boxy thing is real, and the people making the claim need to prove it, my guess is it's the change in directivity from omni at low frequencies to forward radiating when the baffle takes over. Are cardioid speakers also perceived as boxy, when people cannot see the box.

This is what I had thought boxy meant, as in sounds like you are listening in a box hence no spaciousness and that woofer range is dominant. It can derive from poor positioning of a speaker in a box or just a crap design (big directivity change, big step towards the low end)

 

[kemmler3D]

I can't say what boxy sound actually is, but my assumption is that it either has to do with resonances, or it's a way of describing a dispersion pattern that box speakers don't have, implicitly asserting that dipole or omni is a more natural sound.

 

[Arindal]

I am looking for an answer that can specifically be attributed to a "box". I grant that boxed speakers have a different radiating pattern to open baffle / dipole speakers, but in that case it is better to say that you like "dipole sound" so that it is clear that it is the radiating pattern that someone is referring to.

We should separate two things: The directivity pattern caused by the speaker´s geometry and resulting soundfield on one hand, and anything that is happening inside the enclosure. Although it is not a precise term, both can cause phenomena which could be described as ´boxy´.

Note that regarding directivity pattern and resulting reverb tonality, the ´dipole sound´ is not a ´dipole sound´, but similar tonality is achievable by cardioids, larger constant directivity horns, planars, line sources, or even conventional concepts with very large midwoofers as well. What makes the impression ´boxy´ in my understanding, is a baffle step in the upper midrange, which is corrected for linear on-axis response. Meaning below a certain threshold frequency, typically 600 to 800Hz, there is up to +6dB more of indirect sound energy compared to direct sound. Add some reverberant walls in the room and this sounds like the midrange is dominant, longer decaying, distorted, resonant, muddy, just ´boxy´.

Some have said that "box sound" might be caused by the rear wave reflecting off the enclosure and passing through the driver cone or interfering with driver movement, especially since I have heard claims that it can not be measured with a steady-state tone because somehow the cabinet reflection and driver movement are in equilibrium. That sounds like a nonsense made-up argument to me, but I am not so sure.

There are examples of speakers causing this or that internal resonance, or enclosure-induced distortion, which might also be described as ´box sound´. If it is not visible in the frequency response, it usually leaves traces in THD, waterfall plot or alike. Hard to judge if this is the ´box sound´ that some people describe.

 

[RickS]

Speaking of baffles. The most well-known baffle in speaker circles is, guess which one it is:

Spoiler

View attachment 513484

What is a Frequency Response Chart? | Jensen Loudspeakers

www.jensentone.com

Maybe in design circles, but you did remind me of the "infinite" baffle case...

Do some folks really think that an in-wall speaker sounds boxy?

As for the open baffle argument, those speakers are often positioned well away from room boundaries. Would argue that many speakers sound less boxy further out into the room. Seems we cannot get away from the room aspect for this topic either.

 

[Arindal]

As for the open baffle argument, those speakers are often positioned well away from room boundaries. Would argue that many speakers sound less boxy further out into the room.

I would argue that particularly in pro audio and recordings circles, many people attribute ´less boxy´ sound with cardioids and line sources, and they are oftentimes placed very close to a wall. Which could be taken as a hint that the baffle step and imbalance in indirect sound tonality, particularly midrange-heavy reverb and early reflections, is contributing the the ´boxy´ impression.

Interestingly, conventional speakers sound less boxy when placed closer to the wall, as the reflection from the rear wall is cancelling out certain bands, and the boundary effect at least decreases the step in tonal imbalance slightly. The ´boxy effect´ oftentimes is most dominant with reflective walls further away, as the reflections come later and are even more lacking brilliance and treble.

implicitly asserting that dipole or omni is a more natural sound.

If you look solely at the reverb tonality, concepts like dipole and omni do offer serious advantages, as they do not have the typical baffle step between omnidirectional midrange and more or less directional higher bands. This de facto constant directivity leads to unprecedentedly well-balanced, hence ´natural´ reverb. If you hear an acoustic recording on an omni or dipole from a neighboring room, the reverb sounds almost as natural as coming from a live performance IMHO.

This comes at a price, though, of a disadvantageous reflection pattern for stereo reproduction, particularly deteriorated localization for omnis.

 

[RickS]

I can't say what boxy sound actually is, but my assumption is that it either has to do with resonances, or it's a way of describing a dispersion pattern that box speakers don't have, implicitly asserting that dipole or omni is a more natural sound.

While do not want to dismiss the resonance argument offhand, would force me to scratch my head wrt the open baffle argument. Unless you do something to cancel vibrations, simply mounting a woofer on a panel is not special and would argue that any OB design that looks like this...

is likely to have all sorts of panel resonances. Putting a large woofer on a large baffle and driving it hard screams to me to be a more likely source for resonances than a comparable well-braced box design.

Please note! Am not an OB hater but is not a panacea on its own but would avoid overgeneralization of them. Some are much better than others.

 

[Arindal]

Unless you do something to cancel vibrations, simply mounting a woofer on a panel is not special and would argue that any OB design that looks like this...

...

is likely to have all sorts of panel resonances. Putting a large woofer on a large baffle and driving it hard screams to me to be a more likely source for resonances than a comparable well-braced box design.

Resonances within a closed box system which are most likely to be audible, usually involve much lower frequencies, as there must be some mass which is resonating: the air within the enclosure, cabinet walls, you name it. We are talking about frequency bands which in an OB speaker are usually suspect to almost perfect dipole cancellation, so I do not really see a real risk of such resonances reaching the listener´s ear, if they exist at all.

 

[Keith_W]

Let's summarize what we have learnt so far.

Low Q / High Q: not responsible for "boxy" sound. This is because, as @voodooless points out, at low frequencies the Q is dominated by the room, and at high frequencies there is minimal box interaction. I find this line of argument to be convincing, and it's enough for me to dismiss "critical damping" as a factor.

Radiating pattern: I suspect this is what most people mean when they speak of "box sound". I agree that dipoles and omnis sound more "open" (for lack of a better word) than monopoles, but a dipole radiating pattern is not an exclusive quality of panel speakers. And it is better to simply say "I prefer a dipole radiating pattern" rather than "box sound" which I find to be vague.

Cabinet resonances: I would dismiss this one entirely. (1) people who use the term "box sound" seem to imply that all box speakers have it, (2) the majority of a loudspeaker resonances are due to the driver, and (3) it is more likely that a panel speaker will resonate since it is more difficult to brace, as @RickS points out.

Internal cabinet reflections: As @OCA points out, these are easily measurable. Although I should point out that these very early internal reflections are well within the Haas fusion zone and should be masked by the main signal. Anybody disagree with this?

I think this point made by @Arindal is very important:

What makes the impression ´boxy´ in my understanding, is a baffle step in the upper midrange, which is corrected for linear on-axis response. Meaning below a certain threshold frequency, typically 600 to 800Hz, there is up to +6dB more of indirect sound energy compared to direct sound. Add some reverberant walls in the room and this sounds like the midrange is dominant, longer decaying, distorted, resonant, muddy, just ´boxy´.

I seem to recall that Linkwitz made a similar observation. In a conventional monopole loudspeaker, to maintain a flat on-axis response, the sound power has to be tilted downwards, meaning that there is more bass energy in the room. A dipole can have a flat on-axis response AND a flatter sound power curve.

I would like to thank everyone for their thoughtful contributions.

 

[voodooless]

Low Q / High Q: not responsible for "boxy" sound. This is because, as @voodooless points out, at low frequencies the Q is dominated by the room, and at high frequencies there is minimal box interaction. I find this line of argument to be convincing, and it's enough for me to dismiss "critical damping" as a factor.

I didn't say it's not Even when dominated by the room, the basis comes from the speaker, its frequency response, and how it's placed in the room. And the room interaction and general frequency response trends of the two types of speakers are large enough to sound very different in most cases.

 

[Sokel]

"Boxy sound" term comes straightly from these old cheap, thin box speakers with the shouty 200Hz-2kHz response and no meaningful lows or highs.
Pretty much all the cheap, one-driver, full range speakers. You can tell in an instant.

 

[ZolaIII]

@Keith_W I guess it's about ported designs where port tuned fundamental causes refractions which can be observed in FR the Xbox. At least it should be. Open bufle had it's own problems like being unable to achieve good separation. Close bufle compression box is default (for a box) you will find as reference for any driver database. It's driver on it's own and in box and while depth and volume along with bufle size play role regarding Q factor (which again is better to Butterwort) I don't remember anyone mumbling about them in such manner (as there is no Xbox). Closed enclosure can also have fundamental related harmonics that interfere much like one from port but that's for very small boxes (aka headphone cups). Usually this is also thing any half good designer or DIY-er already knows and try to mitigate and limit by design (hardening the box, inserting isolation, counter swing stands and so on) so a good ported design shouldn't have it as much of the problem in the first place.
Other box is the room of course where same apply and usually on quite similar place (small to medium size room with length of 3.5~5m and uper 30 to 40 range for fundamental which matches to typical port tuning used in speakers). You PEQ address it. You might imagine (vividly) what happens when speakers are less then done good in that regard and room fundamental not being taken care of additionally amplify it (a lot). This doesn't mean there aren't greatly done cabinets (for example baltic breach bark vertically pressed from parts) and port tuned (to Fs of woffer it self) speakers but rarely in HiFi and boutique home ones more where it needs to reach higher extreme SPL levels in (quality) PA. At least that's the practice for now.

 

[Arindal]

Low Q / High Q: not responsible for "boxy" sound. This is because, as @voodooless points out, at low frequencies the Q is dominated by the room, and at high frequencies there is minimal box interaction.

That might be true to most of conventional speaker concepts, combining a midwoofer with larger enclosure volumen. In this case, any resonance or deca issues can be more or less traced down to the air in the enclosure (such as a standing wave resonance) or the cabinet.

There are exceptions to this rule, though, particularly with compact midrange drivers in restricted volume as part of a 3-way concept. I have taken part in an experiment comparing different enclosure sizes for the midrange drivers, under anechoic conditions and corrected for on-axis frequency response. The difference is astonishingly huge, but the high/low Q is rather contributing to some kind of more or less warm/full-bodied sound character, not necessarily ´boxy´.

Internal cabinet reflections: As @OCA points out, these are easily measurable. Although I should point out that these very early internal reflections are well within the Haas fusion zone and should be masked by the main signal. Anybody disagree with this?

It depends on how these internal reflections/resonance effects have a chance to leave the enclosure. If the reflex port is the main passway, that might be easily measurable. If it is cabinet walls or diaphragms, not so much. We are talking about wavelengths which make it difficult to time-window the measurement, or separate the measurements of different sources, as crosstalk is an issue even with microphones positioned in the nearfield. Imagine a tower speaker, measuring internally 1m in height - the main vertical standing wave is expected to occur around 170Hz - pretty difficult to measure, if you do not have a waterfall plot with sufficient time window.

In a conventional monopole loudspeaker, to maintain a flat on-axis response, the sound power has to be tilted downwards, meaning that there is more bass energy in the room.

It is not only bass, which could be corrected by EQ as our brain cannot really distinguish direct from reflected bass. It is mainly lower midrange, like 300-800Hz. If due to omnidirectional pattern, the latter is dominant over higher frequency bands in the room, you cannot simply EQ it, because our brain would notice the lack of SPL in the direct sound. That is why broad-baffle speaker, such with large midwoofers or similar concepts, do sound ´thinner´ in the lower midrange and much less ´boxy´, as their baffle step is around 300Hz or even lower.

A dipole can have a flat on-axis response AND a flatter sound power curve.

Omni and dipole are the easiest ways (from mathematical point) to achieve both, as the late S. Linkwitz has pointed out correctly. I would argue, that concepts like (hybrid-) cardioids, line sources, curved planars or very sophisticated multi-way horns/waveguides could achieve the same objective, with almost no downsides.

 

[Neuro]

People keep telling me that they like open baffle speakers because there is no "box sound", but every time I ask them what they mean, they can't tell me. And no, "other speakers sound boxy" is not an answer. I have heard answers such as "it sounds free" or "it doesn't sound constrained", etc. and then I ask them to point at a specific measurement that shows this. Nobody can do it.

I am looking for an answer that can specifically be attributed to a "box". I grant that boxed speakers have a different radiating pattern to open baffle / dipole speakers, but in that case it is better to say that you like "dipole sound" so that it is clear that it is the radiating pattern that someone is referring to. If you have an omnidirectional speaker in a box (e.g. Beolab 90), does it still sound "boxy"? Wouldn't it be better to simply say "monopole sound"?

Today, someone told me that they heard a pair of 18" open baffle subwoofers and said "it doesn't have a box sound, the bass is just there". Subwoofers are omnidirectional, so they can't have a "monopole sound". I asked him what he meant, and I didn't get a very satisfying answer.

Some have said that "box sound" might be caused by the rear wave reflecting off the enclosure and passing through the driver cone or interfering with driver movement, especially since I have heard claims that it can not be measured with a steady-state tone because somehow the cabinet reflection and driver movement are in equilibrium. That sounds like a nonsense made-up argument to me, but I am not so sure.

A more interesting argument is "low Q sound" vs. "high Q sound". Well, boxes can be tuned to different Q's. And drivers for open baffle speakers can be chosen with different Q's. I have been told that "low Q sound" sounds strained and muffled, but I don't have the experience of listening to high Q vs. low Q speakers. And: what measurement would show the effect of Q? Surely all you need to look at is the impulse/step response to see how long the driver rings for?

I am starting to think that "box sound" has no descriptive value, in the same way that "musical" is a useless term. I don't think people agree with what it means, and to the extent that boxes sound different to dipoles, it is better to specifically refer to the radiation pattern ("monopole sound") instead of "box sound". Or "high Q" vs. "low Q" sound. Am I wrong about this?

I am very sensitive to the sound difference between dipole speakers and box speakers.
When comparing my Yamaha NS-1000 (box) and Apogee Duetta (dipole), (both with very good frequency response and very low distorsion) in my large reflex-rich listening room in respective optimal speaker/listening positions, no boomy sounds are identified. There is a prominent difference in the sound in favor of the dipole.
The box speaker lacks space and normal room support from the reflexes. The sound is relatively dull and spaceless.
The sound of the dipole speakers is not diffuse or muddy but rather clearer, lively and more detailed. This occurs with dominant lateral reflexes attenuated by about 8 dB and delayed by about 17 ms.
In smaller rooms the dipole sound can be more muddy and diffuse due to suboptimal reflexes. Optimal reflexes are are decisive.
In smaller listening rooms, the box speaker provides more bass in the listening position relative to the dipole speaker. The dipole provides more destructive interference through phase-reversed sound to the rear. The box speaker has monopole radiation in the bass below a few hundred Hz.

JM

 

[voodooless]

I have taken part in an experiment comparing different enclosure sizes for the midrange drivers, under anechoic conditions and corrected for on-axis frequency response. The difference is astonishingly huge, but the high/low Q is rather contributing to some kind of more or less warm/full-bodied sound character, not necessarily ´boxy´.

That doesn't make sense. If you correct the frequency response, you also correct the Q. So, most certainly, that is not the issue here. Also, you can probably correct a low-Q system to a high-Q without much penalty. The other way around, not so much: you'll need to apply a significant boost. Fine for purpose-built subwoofer drivers, most certainly not for the average midrange. This is tractor tires on a Ferrari all over again...

 

[Arindal]

That doesn't make sense. If you correct the frequency response, you also correct the Q. So, most certainly, that is not the issue here.

No, you do not, as the whole system is not minimum-phase. As all amplitude-, directivity- or nonlinear-distortion-related differences had been avoided by the test design, I was pretty surprised by the scope of the audible differences, particularly de-facto open baffle vs. high-Q closed-box. But as, mentioned, this is applicable to compact midrange drivers within a 3-way system solely, and the differences are measurable via waterfall plot.

 

[EJ3]

People keep telling me that they like open baffle speakers because there is no "box sound", but every time I ask them what they mean, they can't tell me. And no, "other speakers sound boxy" is not an answer. I have heard answers such as "it sounds free" or "it doesn't sound constrained", etc. and then I ask them to point at a specific measurement that shows this. Nobody can do it.

I am looking for an answer that can specifically be attributed to a "box". I grant that boxed speakers have a different radiating pattern to open baffle / dipole speakers, but in that case it is better to say that you like "dipole sound" so that it is clear that it is the radiating pattern that someone is referring to. If you have an omnidirectional speaker in a box (e.g. Beolab 90), does it still sound "boxy"? Wouldn't it be better to simply say "monopole sound"?

Today, someone told me that they heard a pair of 18" open baffle subwoofers and said "it doesn't have a box sound, the bass is just there". Subwoofers are omnidirectional, so they can't have a "monopole sound". I asked him what he meant, and I didn't get a very satisfying answer.

Some have said that "box sound" might be caused by the rear wave reflecting off the enclosure and passing through the driver cone or interfering with driver movement, especially since I have heard claims that it can not be measured with a steady-state tone because somehow the cabinet reflection and driver movement are in equilibrium. That sounds like a nonsense made-up argument to me, but I am not so sure.

A more interesting argument is "low Q sound" vs. "high Q sound". Well, boxes can be tuned to different Q's. And drivers for open baffle speakers can be chosen with different Q's. I have been told that "low Q sound" sounds strained and muffled, but I don't have the experience of listening to high Q vs. low Q speakers. And: what measurement would show the effect of Q? Surely all you need to look at is the impulse/step response to see how long the driver rings for?

I am starting to think that "box sound" has no descriptive value, in the same way that "musical" is a useless term. I don't think people agree with what it means, and to the extent that boxes sound different to dipoles, it is better to specifically refer to the radiation pattern ("monopole sound") instead of "box sound". Or "high Q" vs. "low Q" sound. Am I wrong about this?

I think that it depends on whether you can see the box or not???

 

[Marc v E]

People keep telling me that they like open baffle speakers because there is no "box sound", but every time I ask them what they mean, they can't tell me. And no, "other speakers sound boxy" is not an answer. I have heard answers such as "it sounds free" or "it doesn't sound constrained", etc. and then I ask them to point at a specific measurement that shows this. Nobody can do it.

I am looking for an answer that can specifically be attributed to a "box". I grant that boxed speakers have a different radiating pattern to open baffle / dipole speakers, but in that case it is better to say that you like "dipole sound" so that it is clear that it is the radiating pattern that someone is referring to. If you have an omnidirectional speaker in a box (e.g. Beolab 90), does it still sound "boxy"? Wouldn't it be better to simply say "monopole sound"?

Today, someone told me that they heard a pair of 18" open baffle subwoofers and said "it doesn't have a box sound, the bass is just there". Subwoofers are omnidirectional, so they can't have a "monopole sound". I asked him what he meant, and I didn't get a very satisfying answer.

Some have said that "box sound" might be caused by the rear wave reflecting off the enclosure and passing through the driver cone or interfering with driver movement, especially since I have heard claims that it can not be measured with a steady-state tone because somehow the cabinet reflection and driver movement are in equilibrium. That sounds like a nonsense made-up argument to me, but I am not so sure.

A more interesting argument is "low Q sound" vs. "high Q sound". Well, boxes can be tuned to different Q's. And drivers for open baffle speakers can be chosen with different Q's. I have been told that "low Q sound" sounds strained and muffled, but I don't have the experience of listening to high Q vs. low Q speakers. And: what measurement would show the effect of Q? Surely all you need to look at is the impulse/step response to see how long the driver rings for?

I am starting to think that "box sound" has no descriptive value, in the same way that "musical" is a useless term. I don't think people agree with what it means, and to the extent that boxes sound different to dipoles, it is better to specifically refer to the radiation pattern ("monopole sound") instead of "box sound". Or "high Q" vs. "low Q" sound. Am I wrong about this?

In hindsight what I described as "box sound" was likely the interaction with the room or insuffient bass drivers or insufficient amplification.

I don't hear it with closed box speakers that have ample bass, amplification and are less then half a meter from the wall.

 

[Flaesh]

"Cult of the Infinitely Baffled" Hear The Bass, Not The Box The definitive online resource for Infinite Baffle subwoofer design Established 1999

Home | "Cult of the Infinitely Baffled" Hear The Bass, Not The Box The definitive online resource for Infinite Baffle subwoofer design Established 1999

Visit our forum at: ibsubwoofers.proboards.com

ibsubwoofers.proboards.com

Radiating pattern: I suspect this is what most people mean when they speak of "box sound". I agree that .. omnis

A thought experiment: suppose we have a box with a box-like sound; ideally with a full-range or coaxial speaker for the sake of clarity. Let's place speaker upwards for omnidirectional horizontal directivity. Will the box-like quality disappear?

Pic from an unknown (for me) source: