What Contributes to "Smoothness" In a Speaker's Sound?

[Waxx]

Smooth is often a very flat response, low distortion (especially on the mid and treble) and a lot of headroom. Wtih flat response i mean no big variations (no high resonance peaks) in the spectrum, that is not necesairy flat neutral (altough it helps. Harbeth speakers are known to be smooth, but they are not neutral, they are tuned to a housecurve.

 

[ksulliva01]

Thanks for all of this input!

Of course not everyone here takes offense at sonic descriptors like "buttery." I get that, and was mostly joking. I like that this forum upholds a sense of objective scientific rigor, it's a great counterpoint to the myriad subjective reviews out there. I like to land somewhere in-between both sides, and see the merit of both, but that's an argument for a different thread.

Note - I'm comparing the CSS Criton 1TD not the 1TDX which was reviewed here. It's the same enclosure and woofer, but a different tweeter and crossover. However, from what I've learned in a few of the previous posts, the tweeter-on-a-flat-baffle approach and it's wider dispersion characteristics esp. around crossover point (which is listed at 1.6K for the Criton) could definitely be contributing to the differences I hear and in combination with the little dip that the DBR62 PIR shows between 1K and 4K could completely account for the differences I'm subjectively hearing.

I deleted the image of my measurement, as I'm not confident in the accuracy of it, and therefore do not feel comfortable contributing that data here, and as @AnalogSteph pointed out, the vertical scale was set incorrectly. When bringing down the vertical scale I see more usable data, looking like a little dip 1K-2K with a peak 2K-4K.

 

[Keith_W]

First of all, I disagree with your description that the sound is "buttery". To me it sounds more like margarine.

In any case, the answer is in there somewhere. I know this sounds like a pain in the arse, but you have to keep looking. If the freq response looks the same, look at the phase response, impulse response, group delay, and step response. If it's still not there, then try to take an anechoic or quasi-anechoic measurement. Then try to measure the directivity. Then measure compression. Take electrical measurements and look at the impedance curve. Along the way you will find many differences between the two speakers you are comparing, some of which may not make one sound more like butter than the other. My hunch is that you will find your answer in one of the amplitude response measurements - if not the axial amplitude response, then off axis, distortion, compression, and so on.

 

[kemmler3D]

First of all, I disagree with your description that the sound is "buttery". To me it sounds more like margarine.

In any case, the answer is in there somewhere. I know this sounds like a pain in the arse, but you have to keep looking. If the freq response looks the same, look at the phase response, impulse response, group delay, and step response. If it's still not there, then try to take an anechoic or quasi-anechoic measurement. Then try to measure the directivity. Then measure compression. Take electrical measurements and look at the impedance curve. Along the way you will find many differences between the two speakers you are comparing, some of which may not make one sound more like butter than the other. My hunch is that you will find your answer in one of the amplitude response measurements - if not the axial amplitude response, then off axis, distortion, compression, and so on.

Yep, I have seen a lot of speakers with similar measurements, but I don't think I have ever seen two speakers that measure so similarly (in every aspect) that I would argue the differences should be inaudible or even notably hard to hear, even among high-end studio monitors there are various deviations like the ones you mention.

 

[MattHooper]

First of all, I disagree with your description that the sound is "buttery". To me it sounds more like margarine.

In any case, the answer is in there somewhere. I know this sounds like a pain in the arse, but you have to keep looking. If the freq response looks the same, look at the phase response, impulse response, group delay, and step response. If it's still not there, then try to take an anechoic or quasi-anechoic measurement. Then try to measure the directivity. Then measure compression. Take electrical measurements and look at the impedance curve. Along the way you will find many differences between the two speakers you are comparing, some of which may not make one sound more like butter than the other. My hunch is that you will find your answer in one of the amplitude response measurements - if not the axial amplitude response, then off axis, distortion, compression, and so on.

This is something I’ve wondered about…

A major characteristic that attracted me to my current loudspeakers (Joseph audio perspective 2), when I auditioned them numerous times and through owning them, is a sense of purity - of extremely clean, pure, hash or grain-free sound. When I compared them to a number of other loudspeakers, when I went back to these ones, the sensation was of removing a slit scrim from the sound: Listening to instruments portrayed through the speakers was sort of like viewing an assortment of colourful rocks and pebbles in an aquarium, but where the aquarium had just been cleaned any vestiges of silt or find suspended particles, so one has a slightly cleaner view of the pebbles and their colours. This particular “ purity of tone and lack of distortion” was remarked upon by almost every reviewer of this line of speakers, and typically also by owners of the brand.

So when I go looking for measurements that might explain it: if I look at the frequency measurements, nothing jumps out. There is the typical bump in the Stereophile measurements, a fairly even mid range, and then a slight rise in the highs starting around 5K (though less so with the in room measurements). so that doesn’t seem to be telling me anything.

Which makes me wonder if the culprit is found in the cumulative spectral decay plot, which is generally very clean. I noticed that another line of loudspeakers I’ve owned (and still own for Home Theatre) had always struck me with a similar type of purity, and they two measured very clean in the spectral decay plot. So could this be the reason for the very clean sound?

The thing is, I seem to remember that Amir (and also Toole?) tends to disregard that plot. I believe on the basis that the frequency response will tell you what you need to know.
If there’s resonances or any issues, they will manifest somewhere in the frequency response. So I’m still a bit baffled.

For instance, I auditioned some Revel speakers a number of times while also auditioning the Joseph speakers. The Revel speakers struck me as very well balanced, and very “ smooth” in terms of even frequency response. But they did not stick out to me the way that Joseph did, with that last bit of “purity” and lack of grain, that I’m trying to describe about the Joseph speakers.

I’ve noted in Stereophile measurements that some of the Revel speakers measure pretty even in terms of frequency response, though (if I’ve interpreted correctly) somewhat less clean in terms of the cumulative spectral decay plot then the Josephs.

Not sure if this is significant or not.

All the above comes with obvious caveats that perhaps there’s some sort of bias factor here on my part in regards to perceiving the cleanliness of sound in the Josephs. But since it is such a strong impression, and so regularly echoed by other listeners, it does make me wonder if there is something to it, and if so, what could it be?

 

[Keith_W]

All the above comes with obvious caveats that perhaps there’s some sort of bias factor here on my part in regards to perceiving the cleanliness of sound in the Josephs. But since it is such a strong impression, and so regularly echoed by other listeners, it does make me wonder if there is something to it, and if so, what could it be?

Re: the spectral decay plot. One would imagine that if one frequency decays slower than another, it would affect tonality, and we all know that the prominence of some frequencies might contribute to "grain". As for whether Toole dismisses its importance, I might have missed it in my reading. I have to ask you for a reference.

 

[MattHooper]

Re: the spectral decay plot. One would imagine that if one frequency decays slower than another, it would affect tonality, and we all know that the prominence of some frequencies might contribute to "grain". As for whether Toole dismisses its importance, I might have missed it in my reading. I have to ask you for a reference.

Thanks.

I’m unsure about Toole, but I know Amir downplays the use of the spectral decay plot, and I thought that was on the basis of what Toole has written about the resonances showing up in the frequency response anyway. But I may be getting this wrong.

 

[kemmler3D]

This is something I’ve wondered about…

A major characteristic that attracted me to my current loudspeakers (Joseph audio perspective 2), when I auditioned them numerous times and through owning them, is a sense of purity - of extremely clean, pure, hash or grain-free sound. When I compared them to a number of other loudspeakers, when I went back to these ones, the sensation was of removing a slit scrim from the sound: Listening to instruments portrayed through the speakers was sort of like viewing an assortment of colourful rocks and pebbles in an aquarium, but where the aquarium had just been cleaned any vestiges of silt or find suspended particles, so one has a slightly cleaner view of the pebbles and their colours. This particular “ purity of tone and lack of distortion” was remarked upon by almost every reviewer of this line of speakers, and typically also by owners of the brand.

So when I go looking for measurements that might explain it: if I look at the frequency measurements, nothing jumps out. There is the typical bump in the Stereophile measurements, a fairly even mid range, and then a slight rise in the highs starting around 5K (though less so with the in room measurements). so that doesn’t seem to be telling me anything.

Which makes me wonder if the culprit is found in the cumulative spectral decay plot, which is generally very clean. I noticed that another line of loudspeakers I’ve owned (and still own for Home Theatre) had always struck me with a similar type of purity, and they two measured very clean in the spectral decay plot. So could this be the reason for the very clean sound?

The thing is, I seem to remember that Amir (and also Toole?) tends to disregard that plot. I believe on the basis that the frequency response will tell you what you need to know.
If there’s resonances or any issues, they will manifest somewhere in the frequency response. So I’m still a bit baffled.

For instance, I auditioned some Revel speakers a number of times while also auditioning the Joseph speakers. The Revel speakers struck me as very well balanced, and very “ smooth” in terms of even frequency response. But they did not stick out to me the way that Joseph did, with that last bit of “purity” and lack of grain, that I’m trying to describe about the Joseph speakers.

I’ve noted in Stereophile measurements that some of the Revel speakers measure pretty even in terms of frequency response, though (if I’ve interpreted correctly) somewhat less clean in terms of the cumulative spectral decay plot then the Josephs.

Not sure if this is significant or not.

All the above comes with obvious caveats that perhaps there’s some sort of bias factor here on my part in regards to perceiving the cleanliness of sound in the Josephs. But since it is such a strong impression, and so regularly echoed by other listeners, it does make me wonder if there is something to it, and if so, what could it be?

Since I know you are not allergic to anecdote, I'll chime in on this.

I put up my QRD diffuser in my office yesterday. The effect on the sound is hard to measure and subtle. The easiest way to see it is in the CSD plot, it's there in the ETC plot but not nearly as obvious as you'd expect. Basically zero effect on steady state FR plot. You can see it in the phase plot, but I find it really hard to figure out what that should sound like. The frontier of the CSD plot did clearly get smoother and overall shorter, though.

Totally subjectively, I would say the audible effect of this (if there really is one and I'm not just imagining it) is something like "purity"... I would maybe call it clarity*. The room itself is really bad, the midrange and treble really does have too long of a decay, and it's still far from perfect, but it does seem to have cleaned up the mids and I think I can hear a bit more clearly now.

*In fact, the C50 plot also improved visibly, so that makes sense.

 

[MattHooper]

Since I know you are not allergic to anecdote, I'll chime in on this.

I put up my QRD diffuser in my office yesterday. The effect on the sound is hard to measure and subtle, the easiest way to see it is in the CSD plot, it's there in the ETC plot but not nearly as obvious as you'd expect. Basically zero effect on steady state FR plot. The frontier of the CSD plot did clearly get smoother and overall shorter, though.

Totally subjectively, I would say the audible effect of this (if there really is one and I'm not just imagining it) is something like "purity"... I would maybe call it clarity. The room itself is really bad, the midrange and treble really does have too long of a decay, and it's still far from perfect, but it does seem to have cleaned up the mids and I think I can hear a bit more clearly now.

Interesting. I can totally see that.

I’ve mentioned that I have a fair amount of latitude with the acoustics in my room, especially in terms of reflectivity, since I can pull thick curtains along as much of the wall or to any points in the wall, I want, as well as having diffusion available.

The more of the walls I cover, it seems to reduce what I think of as “ room hash” added to the sound from reflections, especially in the higher frequencies. So a similar thing happens to what I’ve been trying to describe with the speakers themselves, that sense of a slight coarseness or scrim of white spray over the sound, is reduced so the sound seems more pure and direct, instrumental timber becoming a bit more pure and less homogenized. Not to mention somewhat more precise insight into the acoustics or reverb in the recording itself. (which was actually another thing I remember jumping out at me when I auditioned the speakers. it was like many of the tracks. I was familiar with seem to have more reverb than I’ve heard before. The sound was so pure that it made reverb in recordings take on a sort of pristine, luscious quality in of itself.

 

[Doodski]

Interesting. I can totally see that.

I’ve mentioned that I have a fair amount of latitude with the acoustics in my room, especially in terms of reflectivity, since I can pull thick curtains along as much of the wall or to any points in the wall, I want, as well as having diffusion available.

The more of the walls I cover, it seems to reduce what I think of as “ room hash” added to the sound from reflections, especially in the higher frequencies. So a similar thing happens to what I’ve been trying to describe with the speakers themselves, that sense of a slight coarseness or scrim of white spray over the sound, is reduced so the sound seems more pure and direct, instrumental timber becoming a bit more pure and less homogenized. Not to mention somewhat more precise insight into the acoustics or reverb in the recording itself. (which was actually another thing I remember jumping out at me when I auditioned the speakers. it was like many of the tracks. I was familiar with seem to have more reverb than I’ve heard before. The sound was so pure that it made reverb in recordings take on a sort of pristine, luscious quality in of itself.

I agree. Upon installing 2, 6 foot (2m) high and 3 feet (1m) wide sound absorbing panels the room took on a warmer direct sound. The imaging was improved and all the stuff I enjoy like horns, drums, cymbals etc sounded much better. Even talking in the room was a better experience.

 

[MattHooper]

I agree. Upon installing 2, 6 foot (2m) high and 3 feet (1m) wide sound absorbing panels the room took on a warmer direct sound. The imaging was improved and all the stuff I enjoy like horns, drums, cymbals etc sounded much better. Even talking in the room was a better experience.

Exactly. Cymbals in particular are always a good tell.

I’ll always remember how John Atkinson described evaluating a loudspeaker, and mentioning the sound of a high hat. He mentioned that it clearly sounded like wood stick hitting a metallic surface, rather than merely “ modulated bursts of white noise.”

I found that so insightful, because it is so descriptive of what I often hear. Especially if I think of classic disco recordings, you can hear the “ pea soup, pea soup” rhythm of the high hat, and you know it’s a high hat and intellectually. But if you listen closely very often, it really does sound much more crude and diminished. it really could be modulated births of white noise versus a real high hat. When you hear a real high hat near you or play one, the clarity and temporal complexity is just on another level then what often comes through and recordings and sound systems. So for quite a while when I have auditioned different systems, I pay attention to high hats and close my eyes and ask myself “ does this sound just like white burst of noise or does this really sound like a high hat?”

And I often find that, as I’ve described before, drum cymbals , such as ride symbols, rarely come across through stereo systems as the large resonating metal discs they are. They tend to be reduced to tiny points or bursts of high frequency noise, squeezing its way through a tweeter, when compared to the real thing.

But on the topic of room hash/reflections… yeah I find it too much room reflection and high frequencies adds a sort of scrim of hash or white noise that is especially evident and symbols where they all take on that similar “ whitened hashy” tonality.

As soon as I take steps to hear more directly into the recording, that white scrim is wiped away, and as a drummer taps different cymbals they become smoother and more revealing of the tonal and tmbral nuances between different cymbals.

As it happens, I tend to find something similar when I compare a digital recording to the same recording I have on vinyl. There is a slight texture and bur to the sound that is absent in the digital, making for a cleaner smoother more timbrally nuanced sound - drum symbols being one of the first tell.

 

[Doodski]

Exactly. Cymbals in particular are always a good tell.

I’ll always remember how John Atkinson described evaluating a loudspeaker, and mentioning the sound of a high hat. He mentioned that it clearly sounded like wood stick hitting a metallic surface, rather than merely “ modulated bursts of white noise.”

I found that so insightful, because it is so descriptive of what I often hear. Especially if I think of classic disco recordings, you can hear the “ pea soup, pea soup” rhythm of the high hat, and you know it’s a high hat and intellectually. But if you listen closely very often, it really does sound much more crude and diminished. it really could be modulated births of white noise versus a real high hat. When you hear a real high hat near you or play one, the clarity and temporal complexity is just on another level then what often comes through and recordings and sound systems. So for quite a while when I have auditioned different systems, I pay attention to high hats and close my eyes and ask myself “ does this sound just like white burst of noise or does this really sound like a high hat?”

And I often find that, as I’ve described before, drum cymbals , such as ride symbols, rarely come across through stereo systems as the large resonating metal discs they are. They tend to be reduced to tiny points or bursts of high frequency noise, squeezing its way through a tweeter, when compared to the real thing.

But on the topic of room hash/reflections… yeah I find it too much room reflection and high frequencies adds a sort of scrim of hash or white noise that is especially evident and symbols where they all take on that similar “ whitened hashy” tonality.

As soon as I take steps to hear more directly into the recording, that white scrim is wiped away, and as a drummer taps different cymbals they become smoother and more revealing of the tonal and tmbral nuances between different cymbals.

As it happens, I tend to find something similar when I compare a digital recording to the same recording I have on vinyl. There is a slight texture and bur to the sound that is absent in the digital, making for a cleaner smoother more timbrally nuanced sound - drum symbols being one of the first tell.

I really go for cymbals and their crisp sound. Hence I PEQ the death out of stuff in effort to get some semblance of a live sound.

 

[Keith_W]

I put up my QRD diffuser in my office yesterday. The effect on the sound is hard to measure and subtle. The easiest way to see it is in the CSD plot, it's there in the ETC plot but not nearly as obvious as you'd expect. Basically zero effect on steady state FR plot. You can see it in the phase plot, but I find it really hard to figure out what that should sound like. The frontier of the CSD plot did clearly get smoother and overall shorter, though.

I did not want to rain on your parade in your other thread, but Toole did point out that the perceptual effect of several mini-reflections may be the same as a large reflection despite an improvement in the ETC. Chapter 7.3.2 in his book.

 

[MattHooper]

I really go for cymbals and their crisp sound. Hence I PEQ the death out of stuff in effort to get some semblance of a live sound.
View attachment 391734

Your EQ images always crack me up. What are you even doing on this forum?

It reminds me of the EQ that I have to readjust after my son would be driving the car listening to his rap stuff. Massive smile EQ, but of course, with the bass even more boosted.


“ not that there’s anything wrong with that”

 

[Doodski]

Your EQ images always crack me up. What are you even doing on this forum?

It reminds me of the EQ that I have to readjust after my son would be driving the car listening to his rap stuff. Massive smile EQ, but of course, with the bass even more boosted.


“ not that there’s anything wrong with that”

rflmao... My ears have always loved lots of top end. It's not a injury or stuff like that.

 

[DanielT]

I have absolutely nothing against subjective descriptions of how one experiences whatever it may be; sound, food, wine and so on.
If you and your friend are served the same dish at a restaurant and your friend tastes the food first and says it was too little salted. He salts his food. Would you, then, before tasting your food, salt as much as your friend? Or would you EQ 2dB here or there just because your friend does it with the same type of speakers you have (as a thought experiment, with the same type of music and the same listening room).

TS likes the sound from speaker X, in his listening room with those conditions, more than speaker Y. There is nothing strange about that.
Another person may have a different sound taste and instead like speaker Y better under the same conditions. It's good that we don't all like the exact same thing. It would be a sad world if that were the case, that we liked exactly the same that is.
In addition, if TS changes the conditions (more/less furniture, carpets, various acoustic fixing stuff for example) then maybe X is perceived to sound better than Y.

I throw a joker into the deck. TS has a class D amplifier. Maybe it is load dependent and thus, with TS two different speakers speakers, that can create an FR that causes audible differences?
It's not the first thing I'd guess to be an explanation, but I'm throwing it into play anyway.

 

[kemmler3D]

I did not want to rain on your parade in your other thread, but Toole did point out that the perceptual effect of several mini-reflections may be the same as a large reflection despite an improvement in the ETC. Chapter 7.3.2 in his book.

Hmm, I should read the book sometime.

And when I get a chance I'm also going to do some low-grade binaural recordings with my MiniDSP EARS jig so we can all compare the before / after subjectively. I *think* there is a subjective improvement, but of course real-life A/B testing is impossible. I really do feel the sound is less boomy in here, but I could easily convince myself of that without it being the case.

 

[Basilderat]

This is something I’ve wondered about…

A major characteristic that attracted me to my current loudspeakers (Joseph audio perspective 2), when I auditioned them numerous times and through owning them, is a sense of purity - of extremely clean, pure, hash or grain-free sound. When I compared them to a number of other loudspeakers, when I went back to these ones, the sensation was of removing a slit scrim from the sound: Listening to instruments portrayed through the speakers was sort of like viewing an assortment of colourful rocks and pebbles in an aquarium, but where the aquarium had just been cleaned any vestiges of silt or find suspended particles, so one has a slightly cleaner view of the pebbles and their colours. This particular “ purity of tone and lack of distortion” was remarked upon by almost every reviewer of this line of speakers, and typically also by owners of the brand.

So when I go looking for measurements that might explain it: if I look at the frequency measurements, nothing jumps out. There is the typical bump in the Stereophile measurements, a fairly even mid range, and then a slight rise in the highs starting around 5K (though less so with the in room measurements). so that doesn’t seem to be telling me anything.

Which makes me wonder if the culprit is found in the cumulative spectral decay plot, which is generally very clean. I noticed that another line of loudspeakers I’ve owned (and still own for Home Theatre) had always struck me with a similar type of purity, and they two measured very clean in the spectral decay plot. So could this be the reason for the very clean sound?

The thing is, I seem to remember that Amir (and also Toole?) tends to disregard that plot. I believe on the basis that the frequency response will tell you what you need to know.
If there’s resonances or any issues, they will manifest somewhere in the frequency response. So I’m still a bit baffled.

For instance, I auditioned some Revel speakers a number of times while also auditioning the Joseph speakers. The Revel speakers struck me as very well balanced, and very “ smooth” in terms of even frequency response. But they did not stick out to me the way that Joseph did, with that last bit of “purity” and lack of grain, that I’m trying to describe about the Joseph speakers.

I’ve noted in Stereophile measurements that some of the Revel speakers measure pretty even in terms of frequency response, though (if I’ve interpreted correctly) somewhat less clean in terms of the cumulative spectral decay plot then the Josephs.

Not sure if this is significant or not.

All the above comes with obvious caveats that perhaps there’s some sort of bias factor here on my part in regards to perceiving the cleanliness of sound in the Josephs. But since it is such a strong impression, and so regularly echoed by other listeners, it does make me wonder if there is something to it, and if so, what could it be?

I use both Revel F36 and Focal Aria 926 K2s together at the same time. I prefer the Revels, but it's close. The Revels are "smoother', but certainly don't lack hi-frequency detail. The Focals are a little 'brighter' in the high end with maybe a tad more detail When I want more detail I turn up the Focals in the mix, when I want a smoother overall sound I turn up the Revels. Works for me.

 

[MattHooper]

I use both Revel F36 and Focal Aria 926 K2s together at the same time. I prefer the Revels, but it's close. The Revels are "smoother', but certainly don't lack hi-frequency detail. The Focals are a little 'brighter' in the high end with maybe a tad more detail When I want more detail I turn up the Focals in the mix, when I want a smoother overall sound I turn up the Revels. Works for me.

That sounds like the differences I heard when he auditioned the Revels vs the Focal Kanta and Diablo Utopia. And then that’s always been the sound of the consumer Focal speakers, and they measure tipped up as well.

Adds some nice sparkle for cymbals, acoustic guitar, and all that stuff.

 

[Basilderat]

That sounds like the differences I heard when he auditioned the Revels vs the Focal Kanta and Diablo Utopia. And then that’s always been the sound of the consumer Focal speakers, and they measure tipped up as well.

Adds some nice sparkle for cymbals, acoustic guitar, and all that stuff.

... and violins https://alexiy.nl/eq_chart/