Resolution, speed, do these things really exist?

[ADU]

My understanding, btw, is that THD and SINAD are both older methods of measuring and comparing the distortion on different devices. And they may not be the best methods from a perceptual standpoint.

Stay tuned for more on this subject from people like Earl Geddes, Steve Temme, and others in the field. Because there appears to be more research which is still needed to better understand this subject.

 

[BoredErica]

Not really, no
The bandwidth of a system (frequency response) determines its ability to reproduce input signals within that bandwidth.
If a speaker system has a flat, anechoic bandwidth from 20Hz to 20kHz, then input signals in that range can be reproduced too.

I will grant that headphones are a bit different animal.
Same principles apply though.

How is it different for headphones?

Since "speed" of driver is not really a useful concept, what is the advantage of electrostat or planar magnetic drivers for speakers or headphones if any?

 

[dc655321]

How is it different for headphones?

Since "speed" of driver is not really a useful concept, what is the advantage of electrostat or planar magnetic drivers for speakers or headphones if any?

Headphone frequency response must account for the absence of factors that would otherwise be present if listening to sounds not originating a few centimeters from the ear. For example, there is no “room”, gain contributed by ear structure can be altered or absent, etc.

To be honest, I don’t know what advantages, if any, planars or electro stats may hold. Maybe good marketing stories?

 

[JanesJr1]

Headphone frequency response must account for the absence of factors that would otherwise be present if listening to sounds not originating a few centimeters from the ear. For example, there is no “room”, gain contributed by ear structure can be altered or absent, etc.

To be honest, I don’t know what advantages, if any, planars or electro stats may hold. Maybe good marketing stories?

There's a good discussion of this whole topic on Hydrogen here: https://hydrogenaud.io/index.php?topic=72512.0 It reflects some of the comments above but has additional insights.

 

[Feelas]

Yep you are right. My limited experience just like you said. Planar have more detail on individual parts, dynamics have more sense of detail, dynamism, contrast on bigger picture. Planar tend to sound compressed but more real. Slow and effortless. Dynamics tend to sound fast and more speaker like. Generalization from my experience.

Yet this "certain characteristic" of planars vs dynamic doesn't really lend itself to time-FR (sound decay) diagrams all of the time. Seems more like autosuggestion than anything if cannot be shown that the driver doesn't really stop when it should - remember that people *are* in fact doing FR over time diagrams. What does "compressed" mean in planar terms if there is no visible dynamics compression (which is actually not a perceptual name, but a real phenomenon that can also be measured) in the FR diagrams w/ varying loudness? Just curious.

 

[tuga]

What does it mean that some headphones are more resolving or faster than others? I have read many times that more expensive headphones are more resolving or faster than cheaper ones because they have better drivers. But these have always been subjective impressions, can you measure these things in any way?

(Please don't attack me if I said something wrong. I'm a musician, not an audiophile or a technical expert. I'm just trying to understand certain things. When I asked this question to audiophiles, they said I had to hear these differences. I also read Amir reviews but that didn't help me either. My subjective impressions are completely inconsistent. For example, I thought the Beyerdynamic T1 (which Amir doesn't like) are really technically a bit better than the DT880 or HD600 (more transparency, more precision), but other times I think the cheap Sony MH1C (after some equalization) are so good as Campfire Andromeda or any expensive headphones I've heard, and detail, resolution is mainly dependent on equalization...)

Some distortions may affect how one perceives those sonic qualities like perceived "resolution" or "speed".
Even though it is a matter of perception, those subjective aspects will usually have an obvious correlation with measured performance.

 

[tuga]

This interactive table shows how the exaggeration or attenuation of a particular range in the audio band affects how one perceives sound, illustrating how much deviations from flat in frequency response can affect our perception:

https://alexiy.nl/eq/

 

[bkdc]

“Speed” exists, or you can think of it as the ability of the driver to release energy instantaneously. Instant on, instant off. Ideally, you have a massless transducer which vibrates and delivers the energy and than instantaneously stops vibrating without delivering excess energy. You don’t want the driver to continue vibrating and release extra energy that was not recorded. This results in smeared sound.

Thus, no matter what your frequency response graph, you need another graph related to cumulative spectral decay to see how long the driver continues to “ring” or produce sound after the input signal stops.

This is what the thin near-massless metal foil of a RAAL ribbon tweeter on a speaker does AMAZINGLY WELL and why the tweeter details are so transparent versus tweeters on other high regarded speakers (Revels, etc). This quality is not measured by Amir. And this is why speakers should be HEARD. It's also one reason why Salk, Ascend, and BMR speakers are so highly regarded regardless of how they may measure or not.

 

[Shazb0t]

“Speed” exists, or you can think of it as the ability of the driver to release energy instantaneously. Instant on, instant off. Ideally, you have a massless transducer which vibrates and delivers the energy and than instantaneously stops vibrating without delivering excess energy. You don’t want the driver to continue vibrating and release extra energy that was not recorded. This results in smeared sound.

Thus, no matter what your frequency response graph, you need another graph related to cumulative spectral decay to see how long the driver continues to “ring” or produce sound after the input signal stops.

This is what the thin near-massless metal foil of a RAAL ribbon tweeter on a speaker does AMAZINGLY WELL and why the tweeter details are so transparent versus tweeters on other high regarded speakers (Revels, etc). This quality is not measured by Amir. And this is why speakers should be HEARD. It's also one reason why Salk, Ascend, and BMR speakers are so highly regarded regardless of how they may measure or not.

Are you able to link us to measurements done by others that clearly show this?

 

[bkdc]

Are you able to link us to measurements done by others that clearly show this?

A simple Google search of “cumulative spectral decay” will get you far. Spectral decay is more important at higher frequencies (intuitive since you are dealing with shorter wavelengths) than for lower frequencies.

I’m very appreciative of ASR and the objectivist measurements provided here. These are most important for DACS and amplifiers. When it comes to the greatest source of distortion in the audio chain (speakers, headphones), the frequency response measurements and measurement of listening windows are nice. But speakers and headphones actually need to be listened to. You can have the most “accurate” frequency response curve from a tweeter with a well designed crossover but it can still sound like a smear when placed next to a tweeter with a much better decay. Again, the best example is a high quality near-massless ribbon. Another example would be the ‘boominess’ in a poorly damped enclosure.

 

[Shazb0t]

A simple Google search of “cumulative spectral decay” will get you far.

I know what CSD is. I haven't seen reliable CSD measurements clearly showing the RAAL ribbons doing "AMAZINGLY WELL" compared against high end domes which would explain your assertions. The nature of CSD measurements appears finicky to me. This coming from a RAAL ribbon tower owner. I was hoping you could link me to something definitive?

 

[bkdc]

I know what CSD is. I haven't seen reliable CSD measurements clearly showing the RAAL ribbons doing "AMAZINGLY WELL" compared against high end domes which would explain your assertions. The nature of CSD measurements appears finicky to me. This coming from a RAAL ribbon tower owner. I was hoping you could link me to something definitive?

High end domes also do very very well. I’m not saying ribbons are the end all. I’m only saying that you can have FANTASTIC measurements on frequency response and dispersion and still sound inferior. Wouldn’t it be nice if every speaker maker provided the CSD.

Here’s a great old thread. Somebody must’ve updated the CSD plot of the Ascend Sierra-2EX. I don’t recall the Sierra-2EX existing in 2019 but it is a very very impressive CSD above 2kHz. I don’t know if I’ve seen better but it shames the Revel Performa series. Measurements aren’t everything especially with speakers. Whether you call it “speed” or “transparency”, you don’t want your transducer to hang onto energy.

[Video] Speaker measurements by Danny Richie

I just watched it and I loved it. I was stopping the reproduction to compare with others that I have selected, including my beloved KEF Q100 5.25" coaxial (modded). Forced viewing. Check out the tools of loudspeaker design. Updated with the link...

audiosciencereview.com

 

[Shazb0t]

High end domes also do very very well. I’m not saying ribbons are the end all. I’m only saying that you can have FANTASTIC measurements on frequency response and dispersion and still sound inferior. Wouldn’t it be nice if every speaker maker provided the CSD.

Here’s a great old thread.

[Video] Speaker measurements by Danny Richie

I just watched it and I loved it. I was stopping the reproduction to compare with others that I have selected, including my beloved KEF Q100 5.25" coaxial (modded). Forced viewing. Check out the tools of loudspeaker design. Updated with the link...

audiosciencereview.com

You lost me at Danny Richie, he's not a reliable source of information when it comes to these sorts of discussions. He believes, promotes, and sells all sorts of lay audio woo and snakeoil products. These are poor resolution in-room gated measurements. I'm not sold that high resolution properly done measurements of RAAL ribbon CSD compared against the same measurement setup results from high end dome tweeters (Revel Be, etc.) explains anecdotal claims like "the tweeter details are so transparent versus tweeters on other high regarded speakers (Revels, etc)." The low mass stories told about ribbons, planars, and AMTs claiming better fidelity need more definitive evidence in my opinion. It's clear from speakers measured on this forum with the Klippel that these low mass tweeters can have all sorts of resonant issues; just like domes. The low mass isn't a guarantee of anything.

 

[ahofer]

That low mass stories told about ribbons, planars, and AMTs claiming better fidelity need more definitive evidence in my opinion.

As well as the accusation of higher distortion leveled at ribbons (some discussion of this in another thread from the Mesanovic manufacturer)

 

[Shazb0t]

As well as the accusation of higher distortion leveled at ribbons (some discussion of this in another thread from the Mesanovic manufacturer)

I'm not making a generalized claim about all ribbons. I'm trying to say that both ribbons and domes can and do have resonance and distortion issues. You can see that for yourself in speaker measurements on this very forum. That doesn't jive with the generalized anecdotes waxed poetic about low mass ribbons. That's the point I'm trying to make. Poor resolution in-room gated CSD measurements done by a guy with questionable intent doesn't prove anything. I wouldn't use them to support claims pointing either way.

 

[ahofer]

I'm not making a generalized claim about all ribbons.

Didn't mean to imply that, it's just come up elsewhere.

 

[bkdc]

The Sierra-2EX.

Here’s the highly regarded M106. Great speaker by the way. I’m not espousing any “audiophile” claims of magic and fairy dust. The best answer is a level matched blind audition and which speaker sounds better to the listener.

The greatest source of distortion in the audio chain and the most important in end sound quality is the speaker (and the room), and in comparison, the DAC, digital chain, cables, and amplifier introduces orders of magnitude less distortion to be be almost insignificant in comparison (unless you truly have a crappy amplifier).

Amazing that I can’t find a CSR on the M105 or the M126Be. You would figure manufacturers would just publish this data in the interest of transparency.

 

[NTK]

The low mass stories told about ribbons, planars, and AMTs claiming better fidelity need more definitive evidence in my opinion. It's clear from speakers measured on this forum with the Klippel that these low mass tweeters can have all sorts of resonant issues; just like domes. The low mass isn't a guarantee of anything.

The acoustic waves (pressure waves) generated by the speaker diaphragm are due to the acceleration of the air "particles" driven by the diaphragm. Pressure is force per unit area, and acceleration is moving mass, Mms, divided by the motor force minus the loading from the suspension compliance. Therefore, the mass of the diaphragm only affects the sensitivity (acceleration = Force/mass), not the "speed". See Purifi blog post:

A fast driver needs a light cone. Or does it? - PURIFI

Intuition can play us tricks. Take this simple example: how does a speaker cone move when you try to make it produce a “pulse”? I’m willing to bet quite a few of you will guess it will rapidly move forward and then equally rapidly return to its rest position. And from that it’d be quite... View...

purifi-audio.com

 

[bkdc]

The acoustic waves (pressure waves) generated by the speaker diaphragm are due to the acceleration of the air "particles" driven by the diaphragm. Pressure is force per unit area, and acceleration is moving mass, Mms, divided by the motor force minus the loading from the suspension compliance. Therefore, the mass of the diaphragm only affects the sensitivity (acceleration = Force/mass), not the "speed". See Purifi blog post:

A fast driver needs a light cone. Or does it? - PURIFI

Intuition can play us tricks. Take this simple example: how does a speaker cone move when you try to make it produce a “pulse”? I’m willing to bet quite a few of you will guess it will rapidly move forward and then equally rapidly return to its rest position. And from that it’d be quite... View...

purifi-audio.com

The mass in question is the mass of air being displaced. Not the mass of the driver. Unless you think that we’re talking about conservation of momentum. The ideal driver is still a massless driver which is possible only in theory. Again, we’re talking about a different issue which is spectral decay. You want an INSTANTANEOUS STOP as well as an instantaneous attack. You cannot ignore the stopping. You are looking at energy delivered or as a surrogate the total area under the curve in the Y-Z plane on the CSD graph. For an equivalent peak (amplitude) you want the smallest area under the curve and the shortest time/fastest decay in milliseconds. This yields the CLEANER sound. It smears less over time.

 

[NTK]

The mass in question is the mass of air being displaced. Not the mass of the driver. Unless you think that we’re talking about conservation of momentum. The ideal driver is still a massless driver which is possible only in theory. Again, we’re talking about a different issue which is spectral decay. You want an INSTANTANEOUS STOP as well as an instantaneous attack. You cannot ignore the stopping. You are looking at energy delivered or as a surrogate the total area under the curve in the Y-Z plane on the CSD graph. For an equivalent peak (amplitude) you want the smallest area under the curve and the shortest time/fastest decay in milliseconds. This yields the CLEANER sound.

Since apparently you didn't read the Purifi blog post, I will quote some parts of it here:

Intuition can play us tricks. Take this simple example: how does a speaker cone move when you try to make it produce a “pulse”? I’m willing to bet quite a few of you will guess it will rapidly move forward and then equally rapidly return to its rest position. And from that it’d be quite reasonable to conclude that a heavy cone would be pretty bad at doing this. Quite reasonable but sadly also quite wrong.

Basically, what we’re saying is that fundamentally, moving mass only affects sensitivity, while what folks call “fast” or “slow” amounts to bandwidth. And that does not relate to mass, but mostly to how real cones flex and spring back when you push on them. How much depends on material type and geometry but not directly on mass. A case in point: diamond tweeter domes are fabled for their “speed”, i.e. bandwidth, but as it happens, they are also the heaviest domes in regular use. This is reflected in diamond tweeters’ lower than average efficiency. But even though diamond domes are much heavier than their aluminium counterparts, they’re much, much stiffer too.

There’s little point in elaborating the intricacies of membrane design here because you can simply look at the frequency response graph. If a driver’s response is smooth and continues a long way, it’s ‘fast’.