What makes big speakers sound "big"and smaller ones sound "small"?

[gnarly]

I heard them back to back in the same room. Even when adjusted to the same SPL, the Avantgardes sounded huge. It literally sounds like a "wall of sound". So they do some things like reproduce the sense of scale very well - e.g. you can feel a pipe organ. But they get beaten by the little Rogers when it comes to imaging. The Avantgardes makes the singer sound like she has a 6 foot wide mouth, whereas the Rogers has a more believable image.

I've found the same phenomenon listening to the large unity/synergy horn DIYs I make.
Even with a vocalist having a rock solid stereo image, the apparent width of the singers mouth is wider than with smaller speakers.

What's interesting to me, is when I drop stereo and run summed mono to just one side of the large speakers, the size/width of the voice shrinks down to close to that of one smaller speaker, also run in mono.

So I think the "wide mouth" phenom simply has something to do with stereo and geometric scaling.... more so than the large speakers.

Best solution I've found, to get the joy of big speakers and retain tighter imaging of smaller ones, is add a center speaker that's same as left and right, and use a energy preserving matrix to derive new left, new Right, and Center.
Really works well if you have the space, and home theater screens don't mess with the ability to have an equal center speaker.

 

[PeteLeoni]

Women. Yes I know it's "sexist" It's also very true "Those Bose Acoustimass sound so much bigger than those Klipschorns you had before we were married" Not all people named Amir are good people (-:

Or was Bose's name Amar?

 

[benanders]

I did find this study that found that “Overall, the spatial extent on a line source may influence the perception of it’s width.”.
… The other thing is that wouldn’t the apparent source width of a pair of speaker be the distance between them which would cancel any perception of actual speaker width?

That [linked] experiment was very limited in scope and basically established no conclusions. The sample size was not suitable for analyses of subjective interpretations, IMO. Posed some interesting questions and curious evaluation attempts, though.

For speakers without lateral boundaries (sidewalls), “Apparent Source Width” (ASW) might well be the distance between baffle perimeters, given horizontal reflections should not factor in.
Still, that only addresses (at most) half the line of questioning in this thread - ASH (Height) must factor into speakers’ sound “size.” Based on how line arrays in pro / hifi operate, I suspect ASH is more influential than the laterally constrained beaming theoretically resulting from a horizontal array?
And that’s assuming no one associates “depth” of sound or how “hard” a speaker “hits” as contributing to listeners’ perception of a speaker’s sound “size.”

When a stereo effect factors in, ASW should become unpredictable, depending on source content (i.e., music). And then comes the arguments about soundstage width that’s at least partly influenced from the recording process x mixing/mastering chair, etc.

Still glad to have been made aware of the study, thanks @RobL .

 

[MattHooper]

One thing (I probably mentioned before) about a speaker sounding "big" is that, since I keep an ear to the traits of live sound, my impression of most speakers, especially of the size most audiophiles can afford, is that they sound "pear shaped" in size.

So in reference to live instruments, I find that live instruments have a more consistent sense of size and "mass" and physicality.

But with speakers, by a pear shaped sound profile I mean: sonic objects with a lot of bass can sound "large" and full of mass - e.g. a stand up bass, tuba lower registers, low oboe etc. But as you move up the frequency range, instruments more centred in the midrange and then higher, become unnaturally thinner and reductive, until you reach the treble frequencies where things like the higher notes of piccolos or drum cymbals live. And then you get these tiny, thin, slivery versions of the instrument. And drum cymbals sound like small little "bright spots" being tapped in the soundstage, very thin and lacking body.

Whereas in real life a drum set sounds much more balanced, everything sounds BIG, from the snare, toms to the cymbals which sound like the Big Resonating Metal Discs they actually are. The difference between real drums cymbals up close and those teeny little bright spots of sound squeezed out the average tweeter is, to me, pretty enormous.

Same goes for if you hear real music, a stand up bass, tenor sax, clarinet, piccolo, violin, even the smaller instruments hold up with a sense of body and mass against the others, where on most speaker systems the higher frequency range instruments become these slivers of the real thing (think higher woodwind passages at the back of the orchestra, in the average orchestral reproduction from loudspeakers - they don't sound like "real sized instruments from a distance" but like instruments shrunk to lilliputian size and force).

When I hear speakers that produce something closer to the top to bottom evenness and size of "real sounds" that gets my attention quick!

 

[benanders]

...

But with speakers, by a pear shaped sound profile I mean: sonic objects with a lot of bass can sound "large" and full of mass - e.g. a stand up bass, tuba lower registers, low oboe etc. But as you move up the frequency range, instruments more centred in the midrange and then higher, become unnaturally thinner and reductive, until you reach the treble frequencies where things like the higher notes of piccolos or drum cymbals live. And then you get these tiny, thin, slivery versions of the instrument. And drum cymbals sound like small little "bright spots" being tapped in the soundstage, very thin and lacking body.

…

When I hear speakers that produce something closer to the top to bottom evenness and size of "real sounds" that gets my attention quick!

@MattHooper I think it’s an important distinction to make: do you mean un[electrically]amplified instruments only, or do you consider this a common disparity also between, say, pro speakers projecting live sound vs. hifi kit playback? IOW, do you find processed-through-pro-speakers live music to be guilty of the sound profile imbalance you describe (i.e., speaker dispearity)?
Do folks who've weighed in with opinions on this matter find pro speakers to vary in “size sound” or just loudness/impact?

Comparing live instruments (amplified or not) to in-home playback seems hopeless from a standpoint of how most (all?) music seems to be mixed and mastered, regardless of speaker size / behavior. I’m not sure it’s discernible whether the sound profile imbalance you describe is more due to speakers’ behavior than it is to how music for playback is processed during/post-production.

 

[MattHooper]

@MattHooper I think it’s an important distinction to make: do you mean un[electrically]amplified instruments only, or do you consider this a common disparity also between, say, pro speakers projecting live sound vs. hifi kit playback? IOW, do you find processed-through-pro-speakers live music to be guilty of the sound profile imbalance you describe (i.e., speaker dispearity)?

Yes my reference there is unamplified instruments. Amplified instruments are usually also diminished (e.g. comparing a bass played through an average bass cabinet vs on many audiophile home systems).

Do folks who've weighed in with opinions on this matter find pro speakers to vary in “size sound” or just loudness/impact?

Comparing live instruments (amplified or not) to in-home playback seems hopeless from a standpoint of how most (all?) music seems to be mixed and mastered, regardless of speaker size / behavior. I’m not sure it’s discernible whether the sound profile imbalance you describe is more due to speakers’ behavior than it is to how music for playback is processed during/post-production.

I think it's not just loudspeakers, but the recording, mixing process that diminishes the size as well.

It's hard not to have an intuition that some of this has to do with the radiating size, or real estate of the cones - so you've got more driver size devoted to the lower frequencies and by the time you get to the tweeter you have this teeny, tiny little tweeter through which lots of information is being squeezed out. If that's plausible perhaps it tracks somewhat with my impressions where I felt certain loudspeakers sounded more full top to bottom - the Devore O/96 and the MBL Omnis I owned. One of the first things that struck me on the Devores was that cymbals sounded larger than on other speakers I'd auditioned. That speaker which has a semi wave guide as well as a wide baffle to focus all the treble energy towards the listener (and of course being a two way with a big woofer, you are getting some at the top of the woofer's passband aimed at you too). So I wonder if that had anything to to with it. Likewise with the MBLs, the omni tweeter is larger and radiating from every angle is radiating from more surface space. Cymbals I found sounded more substantial on the MBLs. (I seem to remember Magnepan 3.6s also doing a larger upper octave presentation, but it's been a while, but that would be consistent with my conjecture of larger radiating size for the treble).

Total conjecture and I'm not technically compenent enough to know either way.

 

[Kvalsvoll]

So I wonder if that had anything to to with it.

Likely it did.

Your description of small high-frequency instruments is something I also have noticed. So that had to be fixed. It was possible to at least improve things considerably, so you get body and some size of cymbals and triangles.

 

[Trdat]

I have two unique systems, horn with 15 inch mid woofer vs a standard 3 way bookshelf. The horn system has stereo subs(front left/right corners) time aligned with Audiolense(tri-amp) and the 3 way bookshelf has three subs(front left/right and back left) summed into mono not time aligned though with a correction filter with Audiolense that has time alignment against the mains in the AVR with speaker distance( REW support). I probably will tell them apart in a blind test due to the phenomenan of how the bass evnevelopes and the bass lag(good lag just longer than the stereo, I believe audible) associated with the mono summed bass, possibly 3 subs frequencies arriving at different times to the ear. Of course the stereo subs have more powerful motor structure also allowing you to feel the immediate thump.

Point is that if I take away the subs from both systems, I probably wouldn't tell which system is which, even with such unique radiation patterns of both systems. There have been times when I have been confused which system is playing(even though it takes a few min to turn on the horn system due to the tri-amp nature) before I had my subs so perfectly designed and set up. So I am going to agree with Mitch here, as much as I want to believe that the bigger system sounds larger than life the difference is very small and very hard to discern.

I think this points to exactly how muchof a role subwoofers play in a system and how they can be manipulated to provide a certain sound character to the overall sound.

 

[audiofooled]

For me it's (I) narrow directivity and (II) low distortion and (III) low cutoff freq what makes a speaker sound "big" especially at high SPL levels. Flat FR in any case. When you have low distortion and no "room sound overload" higher SPL is not perceived as louder, the sound just gets bigger. It starts to get felt as louder when distortion kicks in at levels below where the ear starts to distort itself heavily and when the room just adds too much reverb/echo/boom...

I agree and this has been my experience too. Especially If low distortion is maintained beyond the point where you can feel frequencies on your skin which are higher than mid-bass. More direct than reverberant sound, lack of audible distortion and more of a tactile feel somehow combines into perception of "big sound", but not "too loud".

Sorry for quoting an old post, but IME these are really key aspects of the phenomenon, albeit how you get there is not at all easy.

 

[Descartes]

Easy to see why View attachment 345424

Is that the 8361?

 

[Pearljam5000]

Is that the 8361?

Yes

 

[JAJDACT]

Point is that if I take away the subs from both systems, I probably wouldn't tell which system is which, even with such unique radiation patterns of both systems. There have been times when I have been confused which system is playing(even though it takes a few min to turn on the horn system due to the tri-amp nature) before I had my subs so perfectly designed and set up. So I am going to agree with Mitch here, as much as I want to believe that the bigger system sounds larger than life the difference is very small and very hard to discern.

I think this points to exactly how muchof a role subwoofers play in a system and how they can be manipulated to provide a certain sound character to the overall sosound.

This has generally been my experience overall as well. I have dual 15" subs at the front of my room and have always made sure to integrate and time align/eq them to achieve the best response. Between three different towers and 7 different bookshelf/monitors that I have tried,they all sounded "large/big" in my listening space. After the bass response I noticed that dispersion and speaker placement made more of a difference in soundstage size perception to me,rather than the speaker size itself. With that being said I have never listened to very large speakers,like JBL 4347 or Klipsch Cornwall's to really compare to smaller floorstanders or bookshelves.

 

[mhardy6647]

Having owned a pair of Cornwalls (1974 vintage) -- they're not really what I'd consider very large. Compare the Cornwall in the center of this Klipsch brochure photo to the Klipschorns flanking it.



... and, truth be told, Klipschorns are indubitably large but not very large.

The 'classic' Cornwall may be physically large from the perspective of folks used to, e.g., KEF LS-50s, but it's a decidedly unimpressively built, nigh-on flimsy box. They're not very heavy and the enclosure's quite resonant. I don't know if the current Cornie model (Cornwall IV, if memory serves) is better built. I certainly hope so, s they are rather pricey.

The aforementioned 1974 pair I had.



Note the unimpressive quality plywood and equally unimpressive construction values.

Altec's somewhat smaller Valencia, e.g., was much more stoutly built, and heavier. Sorry I don't have a side by side photo of the two loudspeaker models. I dispatched the Cornies to the basement (as shown above!) immediately after acquiring a pair of 846A Valencias and never looked back. Indeed, i fairly quickly sold off the Cornwalls -- and I don't get rid of much.



Something like an Altec A4 is a very large loudspeaker, and quite a capable one, given a proper crossover. Altec, bless their hearts, always seemed to underperform when tasked with developing crossovers for their own drivers.

source: https://www.lansingheritage.org/html/altec/catalogs/1975-pro.htm
I do know folks using semi-domesticated A4s (e.g., with the flanking "wings" removed) in home settings, but the A2 is really a tad bit too much for most domestic situations.

 

[MEGB1262]

I have heard in my life a few really big line arrays, they sound really "big", in a smaller room too big though

 

[benanders]

... Especially If low distortion is maintained beyond the point where you can feel frequencies on your skin which are higher than mid-bass. More direct than reverberant sound, lack of audible distortion and more of a tactile feel somehow combines into perception of "big sound", but not "too loud".

...

My B + I in quote for emphasis.
This too has been my (anecdotal) experience in assembling the line array speakers I use. Increased driver surface area x relatively low excursion for frequencies above ~100 Hz relative to sub SA seems to make music “feel” big. With subs off, I find that frequencies above 100 Hz vibrate clothing below 90 dB, as at a close-proximity or amp’ed live venue where projection is considerable despite SPL not being absurdly high. Folks who don’t like a nearfield live experience might not like the effect of the speakers I use.

That leaves me wondering about potential interaction(s) of how we perceive low driver distortion (for replicated efficient drivers barely moving even at high SPL) x increased likelihood for distortion from early/later reflections x # angles / arrival times of reflections.
Truly tall (line) speakers may generate simultaneous same-frequency reflections from an effectively larger perceptible vertical range which. That should hypothetically increase the number of perceptible reflection iterations at a given listening position by way of the (necessarily) increased number of arrival events, no? If real, could increase in reflective iterations / arrival events result in perception of a larger space, i.e. “sound bigger, kind of an inverse-approach-but-same-effect of using a capably powerful speaker in a bigger room?
I get that it might sound unlikely to a low-distortion / flat FR thought camp, but those characteristics don’t tend to match live (= big sound) performances, either, do they?

And what effect if any should there be, if floor bounce = ceiling bounce (should a speaker ideally span floor-to-ceiling for “big sound”)? Might not matter much if the listener’s ears aren’t equidistant between floor/ceiling in the case of reflections, but, if they were?

The drinking word
in my thinking would
be “reflections”.

 

[audiofooled]

My B + I in quote for emphasis.
This too has been my (anecdotal) experience in assembling the line array speakers I use. Increased driver surface area x relatively low excursion for frequencies above ~100 Hz relative to sub SA seems to make music “feel” big. With subs off, I find that frequencies above 100 Hz vibrate clothing below 90 dB, as at a close-proximity or amp’ed live venue where projection is considerable despite SPL not being absurdly high. Folks who don’t like a nearfield live experience might not like the effect of the speakers I use.

That leaves me wondering about potential interaction(s) of how we perceive low driver distortion (for replicated efficient drivers barely moving even at high SPL) x increased likelihood for distortion from early/later reflections x # angles / arrival times of reflections.
Truly tall (line) speakers may generate simultaneous same-frequency reflections from an effectively larger perceptible vertical range which. That should hypothetically increase the number of perceptible reflection iterations at a given listening position by way of the (necessarily) increased number of arrival events, no? If real, could increase in reflective iterations / arrival events result in perception of a larger space, i.e. “sound bigger, kind of an inverse-approach-but-same-effect of using a capably powerful speaker in a bigger room?
I get that it might sound unlikely to a low-distortion / flat FR thought camp, but those characteristics don’t tend to match live (= big sound) performances, either, do they?

And what effect if any should there be, if floor bounce = ceiling bounce (should a speaker ideally span floor-to-ceiling for “big sound”)? Might not matter much if the listener’s ears aren’t equidistant between floor/ceiling in the case of reflections, but, if they were?

The drinking word
in my thinking would
be “reflections”.

IMO more of the direct sound is what gives the sensation of "big sound", provided that it's far field listening and that the speakers are designed to be able to maintain low distortion at high SPL over a greater distance. Room size is a also a factor. In any room there would be a critical distance where, when you continue to move away from the loudspeakers, SPL wouldn't drop any further because sum from all of the reflections, room modes and reverb would be greater than the direct sound. More info on this can be found here:

3 Acoustics

Early reflections can be detrimental to imaging and clarity, but later ones can provide spatial quality. Loudspeaker radiation pattern and actual setup in room is crucial if you want to hear what's in the recording, rather than your room acoustics. Indeed, if system is capable enough, what you may get is sound field which is as big as your field of view, with clarity and imaging from the direct sound. You may try playing this:

To me there's something about visceral feel of, not only the fundamentals, but also harmonic content of large instruments. More over (in this case), percussive character of the performance. The recording technique is such that no wonder it sounds big. Unlike recordings where double bass is only in the phantom center.

 

[Pearljam5000]

Theoretically
Putting the same but drivers
But in 2 in 2 different cabinets
1 big and 1 small
Would the the bigger cabinet definitely sound bigger than the small cabinet ?

 

[Deleted member 48726]

Theoretically
Putting the same but drivers
But in 2 in 2 different cabinets
1 big and 1 small
Would the the bigger cabinet definitely sound bigger than the small cabinet ?

I'm not sure you could even do that, even theoretically. As the drivers for comparison would need to be identical but the drivers' parameters would be more suitable for either the small or the large cabinet.

 

[ahofer]

Yeah, I think you guys are correct....we first have to say what our definition of "big sound" is, before commenting. Tis so personally subjective.

I have two definitions.

First is spatially. There, big sound comes from reflected energy dominating. Think omnis, or perhaps the more familiar Bose 901's.
If you've ever hear 901's in a very live, empty unfurnished room, the sound is gigantic.

Second is dynamic. There, 'big sound' comes from high linear SPL throughout the spectrum, that includes unclipped uncompressed headroom for peak transients.
The entire spectrum matters, by my definition.

^^this^^

But you say there isn't a measurement for the latter (the former we get in CEA-2034). What about Erin's compression measurements I posted earlier?

I feel like there is a decent working hypothesis for measurement here: Wider radiation in the 100-500Hz range and less compression at higher SPLs, all else equal, should produce a 'bigger' sound. I wonder whether a @Floyd Toole or @Sean Olive has any testing experience with that.

Also the comment about late reflections rather than early reflections made intuitive sense to me. That would be consistent with a concert hall experience.

 

[gnarly]

^^this^^

But you say there isn't a measurement for the latter (the former we get in CEA-2034). What about Erin's compression measurements I posted earlier?

I feel like there is a decent working hypothesis for measurement here: Wider radiation in the 100-500Hz range and less compression at higher SPLs, all else equal, should produce a 'bigger' sound. I wonder whether a @Floyd Toole or @Sean Olive has any testing experience with that.

Also the comment about late reflections rather than early reflections made intuitive sense to me. That would be consistent with a concert hall experience.

Glad my thoughts made sense to you.

It's great to see Erin making compression measurements.
The way I read them, is that they reflect average SPL compression....which is compression at its most severe form.

I think what precedes average SPL compression, is peak SPL compression.
I haven't seen a really good test for measuring peak compression yet. I've been trying very short tone bursts (1-5 cycles), and comparing the line level burst to a microphone captured speaker burst. When line level step increases aren't fully matched by acoustic output increases, I figure peak clipping has set in.

Imo, uncompressed (and unclipped) headroom for peak SPL throughout the spectrum is a fundamental component of dynamic sound.
It's not as important as maintaining average uncompressed SPL of course, but I find peak headroom makes for a refreshingly clean sound at higher SPL.

Indeed, i think when a speaker starts sounding loud, it's likely to be running out of headroom first. And then get even louder sounding as it runs out of basic average response linearity.

"Big sound", as defined by dynamic sound, should never sound loud, imo.