Best speakers at low volume

[bonzo75]

Listen to a bunch and hear what you hear. I definitely hear this differentiation between speakers, and the ones that seem to bring a better "aliveness" even at low volumes are usually (but not always) higher efficiency models.

This could happen due to two reasons:
1. Higher efficiency and sensitivity allows the quieter passages to come through clearly. I do find that inefficient speakers can lose out on details during the quieter orchestral passages if not well driven with good control. Efficient speakers are easy to drive, control is easy, letting you enjoy swings from low to high better.
2. Many efficient speakers can be single or two way, and it is easier to maintain consistency between drivers. So both "ways" can be run at the same uniform sound level. With inefficient, low sensitivity cone speakers, often the woofer seems out of whack with the upper range drivers, causing the listener to adjust the volume for the lowest common denominator, leading the other driver to sound louder.

 

[morricab]

I don't get why any speaker would be better at low volume than another unless they have a tailored response that approximates Fletcher Mulson. That is a characteristic that I certainly would not want. At low volume nothing is stressed so there should be no challenges at all for any decent speaker system.

Rob

I think it is pretty clear why this would be the case. It is not about stress on the system it is about resistance to turn electrical energy into mechanical energy. A high sensitivity speaker is by definition more responsive to a given input signal than a low sensitivity speaker. Once you go down in volume sufficiently there will be a point at which the speaker basically refuses to move when fed a very small signal. The reasons for a given speaker to be less sensitive could be the following: high mass of the driver assemblies relative to magnetic field strength, mechanical resistance due to surround/spider stiffness, crossover losses etc. Once the driving force is less than the static resistance force then that sound will simply not get produced.

No, thinking about all the low level information that supports the feeling of sounding "alive" rather than "dead" it is not that hard to see why some speakers, which continue to supply that information as the signal level goes down, would sound more "alive" at low volumes than other speakers where that information starts to drop off. Magnepans, for example, were notorious for needing to be "woken up" by playing them a bit more on the loud side and not doing very well at low volumes.

Conversely, even though my Acoustats were low sensitivity (around 84dB) they still sounded lively because the main resistance to panel motion was purely a low electrical sensitivity. The membrane still responded even to very small input force and therefore kept the liveliness down to pretty low levels. I think Magnepan's problems was that the wires on the membrane made the mass of the panel too heavy and the magnets were not strong enough and so there was a lot of resistance to motion. There are some other, rare, examples of low sensitivity speakers that can still sound lively at low levels but the majority of low sensitivity speakers simply sound dead at lower levels.

 

[DasguteOhr]

Lowther Acousta and Saba Greencones (cello) requires millivolts to hear good

 

[morricab]

The Horning Eufrodite Ellipse PM65 speakers I now have in house work very well at low volumes. The high sensitivity (98dB) and the TQWT cabinet design leads to getting all kinds of information with out playing loud. The de-whizzered Lowther is their top of the range unit and is incredibly responsive. There is what I like to call "touch" to the sound where little contact sounds get retained that are lost in other speakers.

Good single driver speakers excel at this "touch" aspect (I guess this is why the Songer speakers are now making a splash) and one of the strengths I hear every year in Munich when I listen to Voxativ (although I have other issues with their sound sometimes...depends on the speaker model), Supravox or AER. To a lesser extent Cube Audio and Fostex. Where I don't like single drivers is sometimes the "shout" and I don't like the whizzer cone highs, which sound buzzy and dry to me.

 

[Stefan]

The topic is also interesting to me, and my experience, in line with many of the comments shared, is that higher-sensitivity speakers do better at communicating the essence of a recording at lower volume levels. I am delighted with the Living Voice Avatar range from the UK in this regard.

However, besides the speaker's sensitivity, I think it is worth factoring in the Equal-loudness curve(Fletcher-Muncon curve). In essence, our hearing across the frequency range is not linear, and changes will the loudness levels as well. Specific speakers have a loudness curve designed in their voicing and might be more satisfying at lower sound pressure levels. The Converse is that if a speaker was conceived and designed to be played at higher volumes, it might not do as well when listened to at lower levels.

On a more detailed level, the design of a speaker is highly relevant. Even in the case of the radius of ports flares, there might be a theoretical difference between speakers. Some ports have lower distortion at low sound pressure than higher levels and vice-versa.

 

[morricab]

The topic is also interesting to me, and my experience, in line with many of the comments shared, is that higher-sensitivity speakers do better at communicating the essence of a recording at lower volume levels. I am delighted with the Living Voice Avatar range from the UK in this regard.

However, besides the speaker's sensitivity, I think it is worth factoring in the Equal-loudness curve(Fletcher-Muncher curve). In essence, our hearing across the frequency range is not linear, and changes will the loudness levels as well. Specific speakers have a loudness curve designed in their voicing and might be more satisfying at lower sound pressure levels. The Converse is that if a speaker was conceived and designed to be played at higher volumes, it might not do as well when listened to at lower levels.

On a more detailed level, the design of a speaker is highly relevant. Even in the case of the radius of ports flares, there might be a theoretical difference between speakers. Some ports have lower distortion at low sound pressure than higher levels and vice-versa.

Which speakers are designed using Fletcher-Munson curves to compensate for losses at low levels? The only way I could imagine this could be done is with a fully active speaker and DSP that is programmed with tailoring frequency response curves for different SPLs. This would need to be a real-time algorithm acting on the signal. Also, when you have soft sounds in the presence of loud sounds, what curve to you apply? Would you have to apply the algorithm separately for those sounds or just for the average signal level?

 

[Stefan]

Which speakers are designed using Fletcher-Munson curves to compensate for losses at low levels? The only way I could imagine this could be done is with a fully active speaker and DSP that is programmed with tailoring frequency response curves for different SPLs. This would need to be a real-time algorithm acting on the signal. Also, when you have soft sounds in the presence of loud sounds, what curve to you apply? Would you have to apply the algorithm separately for those sounds or just for the average signal level?

I am hesitant to refer to specific speakers; suffice it to say that many designs with a BBC smilley curve might come to mind. Of course, this could spell less than optimal sound at higher sound levels. However, I think the Equal-Loudness curve does have some bearing on the topic. The Harman studies that tried to develop an "ideal curve" for loudspeakers might not have fully controlled for this factor.

In my reflections and thoughts on the subject, I also am intrigued by the variability in the sound levels at which recordings are mastered. It could impact how we perceive the playback of the recording in our environments.

It might be a complicated topic and therefore is conveniently sidestepped by many. I certainly don't have the answers but it partly explains why we have such richness in views, designs and preferences.

 

[Robh3606]

I think it is pretty clear why this would be the case. It is not about stress on the system it is about resistance to turn electrical energy into mechanical energy. A high sensitivity speaker is by definition more responsive to a given input signal than a low sensitivity speaker. Once you go down in volume sufficiently there will be a point at which the speaker basically refuses to move when fed a very small signal. The reasons for a given speaker to be less sensitive could be the following: high mass of the driver assemblies relative to magnetic field strength, mechanical resistance due to surround/spider stiffness, crossover losses etc. Once the driving force is less than the static resistance force then that sound will simply not get produced.

Hello

Agree to disagree on this point. We are talking low level in room listening.

This is simply never going to happen where a cone driver "sticks" at any input level where we can hear sound in the room. High efficiency drivers tend to have stiffer suspensions by design. For example woofers higher FS so if anything is going to "stick" it would be the HE system getting milliwatts instead of watts into the lower efficiency design.

Rob

 

[morricab]

Hello

Agree to disagree on this point. We are talking low level in room listening.

This is simply never going to happen where a cone driver "sticks" at any input level where we can hear sound in the room. High efficiency drivers tend to have stiffer suspensions by design. For example woofers higher FS so if anything is going to "stick" it would be the HE system getting milliwatts instead of watts into the lower efficiency design.

Rob

No, you have it backwards…the HE speaker only NEEDS mW to move with the same force (ie SPL) as the low sensitivity speaker that NEEDS much more power to move with them same force. Put the same power in and see which one moves with more force…
The lossiness and low efficiency means a less responsive driver to small inputs.

 

[Robh3606]

No, you have it backwards…the HE speaker only NEEDS mW to move with the same force (ie SPL) as the low sensitivity speaker that NEEDS much more power to move with them same force. Put the same power in and see which one moves with more force…
The lossiness and low efficiency means a less responsive driver to small inputs.

No actually I don't. If you are talking equivalent/the same low level SPL where if it's audible in room there will be no "sticking". The power levels you are talking about I frankly have no clue, to be so low that the voice coil doesn't move.

As far as lossiness it's just being reproduced at a lower SPL. There is no evidence that lower level signal detail just disappears. If it's inaudible it is getting masked from room noise. This will also happen with a high efficiency system at the same SPL.

You are hinting at Dynamic Compression which is not going to happen during low level listening.

We are not talking about the same power it's low level in room and at the same SPL . At the same SPL your argument doesn't hold up IMHO.


Rob

 

[morricab]

No actually I don't. If you are talking equivalent/the same low level SPL where if it's audible in room there will be no "sticking". The power levels you are talking about I frankly have no clue, to be so low that the voice coil doesn't move.

As far as lossiness it's just being reproduced at a lower SPL. There is no evidence that lower level signal detail just disappears. If it's inaudible it is getting masked from room noise. This will also happen with a high efficiency system at the same SPL.

You are hinting at Dynamic Compression which is not going to happen during low level listening.

We are not talking about the same power it's low level in room and at the same SPL . At the same SPL your argument doesn't hold up IMHO.


Rob

That's just the point, low sensitivity speakers start to lose those low level sounds and it is due to the lossiness of the driver (electrical and mechanical lossiness).

It is not just being reproduced at a lower level...at some point those very low level signals will drop off completely...either the driver fails to reproduce them or they are at a level below audibility...thus the loss in "aliveness".

I am not hinting at dynamic compression...although that also happens at lower SPL than with a high sensitivity speaker. So, a low sensitivity speaker is capped at both ends of the SPL spectrum.

The point is that at low SPL the high sensitivity speaker is still reproducing those subtle dynamic shifts and ambience cues that lead to an alive perception. At the same nominal low SPL, the low sensitivity speaker will not capture those things correctly and I would argue that is due to the inherent inefficiency of the speaker itself and the various parameters that contribute to that low sensitivity.

 

[Lagonda]

You guys should stop using those shitty/weak SET amps you ate so fond of ! Problem solved !

 

[morricab]

You guys should stop using those shitty/weak SET amps you ate so fond of ! Problem solved !

No, because then you have dynamic compression issues with low sensitivity speakers...power doesn't overcome this issue.

 

[bonzo75]

You guys should stop using those shitty/weak SET amps you ate so fond of ! Problem solved !

Or get better speakers to go with the low watt amps. You do understand that in this analogy the big SS amps are not big in a powerful muscular beach body way, but in an obese way.

 

[Lagonda]

Or get better speakers to go with the low watt amps. You do understand that in this analogy the big SS amps are not big in a powerful muscular beach body way, but in an obese way.

I hear quick, nimble, and ripped from my SS behemoths, no excess fat here ! The plumpness is only heard when i stick some tubes in the signal chain occasionally !

 

[Robh3606]

That's just the point, low sensitivity speakers start to lose those low level sounds and it is due to the lossiness of the driver (electrical and mechanical lossiness).

Your are confusing driver losses from mechanical issues which effects efficiency with resolution. One effects driver SPL and not the driver resolution. Just because it's at a lower level doesn't mean a loss of program material/detail always occurs.

I am not debating the advantages of HE systems at normal listening levels. We are talking about low level background where the advantages are not going to be as apparent if audible at all.

The point is that at low SPL the high sensitivity speaker is still reproducing those subtle dynamic shifts and ambience cues that lead to an alive perception. At the same nominal low SPL, the low sensitivity speaker will not capture those things correctly and I would argue that is due to the inherent inefficiency of the speaker itself and the various parameters that contribute to that low sensitivity.

And whats the point if you can't hear them??

Low level listening where the perceived frequency response is shifted because of Flet/Mul. It' is never going to sound right without a tone control or loudness switch.

Low level listening where dynamic range of the source material can make lower levels in-audible or compromised from room noise. The best bet would be to compress the source material.

Rob

 

[morricab]

Your are confusing driver losses from mechanical issues which effects efficiency with resolution. One effects driver SPL and not the driver resolution. Just because it's at a lower level doesn't mean a loss of program material/detail always occurs.

I am not debating the advantages of HE systems at normal listening levels. We are talking about low level background where the advantages are not going to be as apparent if audible at all.



And whats the point if you can't hear them??

Low level listening where the perceived frequency response is shifted because of Flet/Mul. It' is never going to sound right without a tone control or loudness switch.

Low level listening where dynamic range of the source material can make lower levels in-audible or compromised from room noise. The best bet would be to compress the source material.

Rob

Of course it affects resolution...what do you think resolution is? I didn't say always...I said most.

Low level background signals is exactly where the advantage of high sensitivity speakers exists...what do you think contributes to the sense of aliveness?? Not just macro dynamic swings but the down in the weeds shifts that you will hear live that you won't hear with most low sensitivity speakers.

Of course the tonal balance shifts as you go down in level without compensation but there are further losses that skew things further away from where you want to be nominally. The best speakers preserve the signal better as it goes down in level...the F-M curve is...just always there. You can compensate or not...most people don't because, as I outlined above, it is not trivial to do it correctly at all SPL levels dynamically.

Well, most material is compressed. However, the best sounding material has relatively little but has to be played on a system that can handle well both ends of the SPL spectrum. Low sensitivity speakers with high powered amps can handle (to a point) the upper SPL range but do poorly on the lower end ...so they basically NEED compressed sources to bring up the low stuff and down the really big peaks (because they will also dynamically compress that peaks worse than a high sensitivity speaker). Most pop/rock is heavily compressed to work with very dynamically compromised systems (boom boxs, JBL pills, ear buds, car stereos etc.). Of course then you will hear lower level details (if there actually any in the music) because they have been boosted and you won't dynamically compress too badly because the high levels have been squeezed as well.

I thought we are talking about high end systems here that should thrive with relatively uncompressed recordings and show just how much damage is done with heavy compression?? Now you are talking about "The best bet would be to compress the source material". Do you work for a record company?

One of the most bizarre speakers I ever had for this kind of issue was a pair of Dynaudio Contour 1.8 MkII floorstanding speakers. With typically pop/rock they sounded quite good was long as you kept the average SPL range from about 75-90dB. If you got below that the sound because pretty lifeless and if you got above that they would start to sound compressed. If you had compressed Jazz all was still ok but with wider dyanmic range jazz and especially classical you could not really set them a good level to satisfy either the low level losses in lower SPL passages or the dynamic compression heard when one turned up the low SPL passages and then didn't turn it back down for the louder SPL passages. So, gain riding was necessary for dynamic source material...I was acting as my own human compressor! Once I started to broaden my music choice these speakers had to go.

 

[Robh3606]

LOL well we seem to be talking past one another! Did you notice what's in my avatar?? I have 5 current systems in use 4 of which use horns/waveguides on the HF drivers one runs at 98 another at 95 1 watt i meter. The 5th is cone and dome a pair of Revel Performa F 206's. I am well aware what the perceived differences are between HE vs typical systems.

I thought we are talking about high end systems here that should thrive with relatively uncompressed recordings and show just how much damage is done with heavy compression?? Now you are talking about "The best bet would be to compress the source material". Do you work for a record company?

Why are you taking a suggestion of compressing wide dynamic range material when listened to low level where it would benefit and make it more ineligible sound like a terrible idea??? Are you not familiar to what Late Night Mode does in an HT system?? That and tone controls and most of the issues with low SPL level play just go away.

One of the most bizarre speakers I ever had for this kind of issue was a pair of Dynaudio Contour 1.8 MkII floorstanding speakers. With typically pop/rock they sounded quite good was long as you kept the average SPL range from about 75-90dB. If you got below that the sound because pretty lifeless and if you got above that they would start to sound compressed. If you had compressed Jazz all was still ok but with wider dyanmic range jazz and especially classical you could not really set them a good level to satisfy either the low level losses in lower SPL passages or the dynamic compression heard when one turned up the low SPL passages and then didn't turn it back down for the louder SPL passages. So, gain riding was necessary for dynamic source material...I was acting as my own human compressor! Once I started to broaden my music choice these speakers had to go.

There isn't a speaker system out there that doesn't require an SPL level that makes them come alive. It's driven by the source material. You tend to play at levels that sound correct to the recording.

What is the sensitivity the 80's??? Lousy design if that's the case never heard a pair

As far as high end systems who listens where the levels are wrong as in to low or too loud for the recorded material??

I sure as hell don't I use appropriate levels. Not even sure any of my rigs would even qualify for high end on this forum.

Rob

 

[gleeds]

I'm not sure I agree that efficiency is not a positive factor when a focus is performance at lower listening levels. The Fletcher Munson curve certainly comes into play but is likely not much of a factor at, say, 60db. In my experience, the more coherent a speaker is the more realistic it sounds at lower levels. TAD with its concentric driver, is such a speaker. Devore may also be good in this regard, although I have yet to hear them.

Two speakers outside the mainstream may fit the bill nicely are both field coils, one from Wolf Von Langa, the "Son" and the other from Songer Audio. The field coil power supply in the Son allows you to control the magnetic field of the woofer and therefore modify the level of woofer damping to get the bass dialed in to suit your room and may be useful for tuning at lower listening levels. Both speakers I referenced are efficient, 92 and 93 dB, respectively.

Ricco, good luck!

 

[jbrrp1]

The TAD's are not efficient speakers, but I agree that they possess this trait of still sounding good at lower volumes, more "complete" than many. Maybe that is because of that concentric coherence. I know Reference 3A speakers stand out for me in lower volume use, with their low-to-mid 90's efficiency and no crossover on the mid/bass drivers.