Between measurements and sound

[DNCAgain]

I've seen some people claim that current acoustic measurement systems cannot reflect changes in perceived sound quality, and that only in-person listening tests can reveal the true sound (perhaps they are unaware of psychoacoustics). They also argue that even scientific measurements on ASR and from other bloggers can lose objectivity due to commercialization.(However, as far as I know, ASR does not accept commercial involvement.)

In fact, for most high-performing speakers measured on ASR, the differences in data are indeed minimal. Yet, listening experiences can vary significantly.

So, what is the truth? How exactly do scientific measurement data connect to our listening experience? Is the existing system perfect? Why can't "listening tests" accurately reveal a speaker's true capabilities?

I hope to get recommendations for a few must-read posts so I can thoroughly study the topic of data versus listening experience.

 

[Sernyl]

I'm not an expert but, for supposedly equivalent measurements, IMHO the listening experience depends of :
- Directivity lobes at different frequencies (see the increasing interest to cardioid patterns).
- Your room parameters: they are unique.
- Placement (distances, toe-in ...etc).
- Individual tastes, including genre of music.

(back in my hole)

 

[DNCAgain]

I'm not an expert but, for supposedly equivalent measurements, IMHO the listening experience depends of :
- Directivity lobes at different frequencies (see the increasing interest to cardioid patterns).
- Your room parameters: they are unique.
- Placement (distances, toe-in ...etc).
- Individual tastes, including genre of music.

(back in my hole)

What about IEMs?

 

[sigbergaudio]

I guess this recent thread may be of interest to you:

What role should listening play in speaker design

Note from staff: This thread has been split from another thread and best effort was made to preserve the flow. When we are talking whether a speaker has flat on axis response and well behaved off-axis, Andrew Jones (like other designers) leaves room for his own judgement over using...

www.audiosciencereview.com

And those wanting to discuss the topic is also typically referred to this thread:

Master Thread: Are Measurements Everything or Nothing?

Seems like so many review threads get challenged with: 1. Measurements are not everything. 2. You all never listen. 3. I trust my ears, not graphs. 4. I don't listen to graphs. I listen to music. 5. You all must not listen to music at all. 6. Why don't you all buy the best SINAD gear? 7...

www.audiosciencereview.com

 

[DNCAgain]

I guess this recent thread may be of interest to you:

What role should listening play in speaker design

Note from staff: This thread has been split from another thread and best effort was made to preserve the flow. When we are talking whether a speaker has flat on axis response and well behaved off-axis, Andrew Jones (like other designers) leaves room for his own judgement over using...

www.audiosciencereview.com

And those wanting to discuss the topic is also typically referred to this thread:

Master Thread: Are Measurements Everything or Nothing?

Seems like so many review threads get challenged with: 1. Measurements are not everything. 2. You all never listen. 3. I trust my ears, not graphs. 4. I don't listen to graphs. I listen to music. 5. You all must not listen to music at all. 6. Why don't you all buy the best SINAD gear? 7...

www.audiosciencereview.com

Thanks for sharing!

 

[Super Panda]

What about IEMs?

While measurements capture the transducer's output, they cannot perfectly predict the final pressure at a specific eardrum.
Measurement rigs use standardized couplers, but human ear canals vary in volume, shape, and tissue compliance. This changes the "load" on the driver which alters the actual frequency response. Also, the distance between the tip and the eardrum dictates the frequency of the half-wave resonance that shifts depending on insertion depth.

Speaker measurements describe the object (the speaker) and are highly predictive, while headphone/IEM measurements describe the interaction between the object and a "standard" human and are highly variable.

 

[DNCAgain]

While measurements capture the transducer's output, they cannot perfectly predict the final pressure at a specific eardrum.
Measurement rigs use standardized couplers, but human ear canals vary in volume, shape, and tissue compliance. This changes the "load" on the driver which alters the actual frequency response. Also, the distance between the tip and the eardrum dictates the frequency of the half-wave resonance that shifts depending on insertion depth.

Speaker measurements describe the object (the speaker) and are highly predictive, while headphone/IEM measurements describe the interaction between the object and a "standard" human and are highly variable.

So, for headphones and IEMs, the actual frequency response does vary due to differences in individuals and usage methods, right? Are there any studies that indicate the extent of this variation? And have manufacturers made any efforts to reduce them?

 

[Matt_Holland]

I've seen some people claim that current acoustic measurement systems cannot reflect changes in perceived sound quality, and that only in-person listening tests can reveal the true sound (perhaps they are unaware of psychoacoustics). They also argue that even scientific measurements on ASR and from other bloggers can lose objectivity due to commercialization.(However, as far as I know, ASR does not accept commercial involvement.)

In fact, for most high-performing speakers measured on ASR, the differences in data are indeed minimal. Yet, listening experiences can vary significantly.

So, what is the truth? How exactly do scientific measurement data connect to our listening experience? Is the existing system perfect? Why can't "listening tests" accurately reveal a speaker's true capabilities?

I hope to get recommendations for a few must-read posts so I can thoroughly study the topic of data versus listening experience.

The absolutely essential read:

It’s very important, I think, to acknowledge the best data and scientific understanding of today before entertaining claims like those you allude to.

 

[TimVG]

Good measuring loudspeakers cannot be guaranteed to sound good always due to external factors. It depends on what you play through them, and where you play them.
However, bad measuring loudspeakers are guaranteed to sound bad under all circumstances

 

[DNCAgain]

The absolutely essential read:
View attachment 522281

It’s very important, I think, to acknowledge the best data and scientific understanding of today before entertaining claims like those you allude to.

Thanks!

 

[Matt_Holland]

Good measuring loudspeakers cannot be guaranteed to sound good always due to external factors. It depends on what you play through them, and where you play them.
However, bad measuring loudspeakers are guaranteed to sound bad under all circumstances

I disagree on the last part. In a room with good acoustics and an optimal setup, a pair (or more) of bad measuring loudspeakers could sound good with the right program material. I think this is exactly how bad measuring speakers get good reviews when using audiophile music to subjectively assess them. And we know that 2-channels and stereo recordings masks speaker flaws.

 

[Super Panda]

So, for headphones and IEMs, the actual frequency response does vary due to differences in individuals and usage methods, right?

Correct.

Are there any studies that indicate the extent of this variation?

I'd recommend checking out the latest headphones.com videos on Youtube. IMO they do a decent job of explaining magnitude variations.

And have manufacturers made any efforts to reduce them?

At least Bose and Apple have, as far as I'm aware.

 

[TimVG]

I disagree on the last part. In a room with good acoustics and an optimal setup, a pair (or more) of bad measuring loudspeakers could sound good with the right program material. I think this is exactly how bad measuring speakers get good reviews when using audiophile music to subjectively assess them. And we know that 2-channels and stereo recordings masks speaker flaws.

If your program material happens to have an inverse response compared to your loudspeaker - sure, but that's looking for a needle in a haystack.
Bad measuring speakers get good reviews simply because it is part of a business model.

 

[DNCAgain]

Correct.

I'd recommend checking out the latest headphones.com videos on Youtube. IMO they do a decent job of explaining magnitude variations.

At least Bose and Apple have, as far as I'm aware.

Thanks for that,but l can't have access to any of the YouTube videos;(

 

[Super Panda]

Thanks for that,but l can't have access to any of the YouTube videos;(

You might find some written content on their website and online communities.

 

[RandomEar]

Just to point it out: This discussion focuses on transducers like speakers, headphones and IEMs, which makes sense because those have a far bigger influence on the sound than other components. However, the typical audiophile claims also include the same sentiment ("I can hear it but the measurements don't show it") for DACs, amps or even individual parts such as capacitors and cables. Clearly, the claims about cables sounding different are at best clueless. So the fact that those claims exist does not in itself prove that there is any difference or that measurements are missing anything.

However, as pointed out by fellow members, differences in speakers are far larger than those in other components and due to qualities like directivity, the interaction with the room (or your ear in case of IEMs) is quite large and will affect the sound differently for pretty much any listener, because no two rooms (or ears) are the same.

I also disagree somewhat that the differences between high performing speakers measured on ASR are small. It's difficult to specify what exactly small means in this context. Differences especially in the bass extension are still common and bass is super important for our listening pleasure. Equally, a rather small dip of the directivity in the crossover region will often be in the frequency region of speech, where our ears are most sensitive. This can easily make or break the subjective impression of a speaker.

 

[OCA]

Great speakers are shoebox shaped designs with plastic waveguides that achieve strong directivity performance when measured with a Klippel near-field scanner. Yet many major manufacturers continue to over engineer their products to sustain the “snake oil” side of the industry, like the example below

 

[DNCAgain]

Just to point it out: This discussion focuses on transducers like speakers, headphones and IEMs, which makes sense because those have a far bigger influence on the sound than other components. However, the typical audiophile claims also include the same sentiment ("I can hear it but the measurements don't show it") for DACs, amps or even individual parts such as capacitors and cables. Clearly, the claims about cables sounding different are at best clueless. So the fact that those claims exist does not in itself prove that there is any difference or that measurements are missing anything.

However, as pointed out by fellow members, differences in speakers are far larger than those in other components and due to qualities like directivity, the interaction with the room (or your ear in case of IEMs) is quite large and will affect the sound differently for pretty much any listener, because no two rooms (or ears) are the same.

I also disagree somewhat that the differences between high performing speakers measured on ASR are small. It's difficult to specify what exactly small means in this context. Differences especially in the bass extension are still common and bass is super important for our listening pleasure. Equally, a rather small dip of the directivity in the crossover region will often be in the frequency region of speech, where our ears are most sensitive. This can easily make or break the subjective impression of a speaker.

Great speakers are shoebox shaped designs with plastic waveguides that achieve strong directivity performance when measured with a Klippel near-field scanner. Yet many major manufacturers continue to over engineer their products to sustain the “snake oil” side of the industry, like the example below
View attachment 522304

I've heard of a bass-reflex design called the "labyrinth" type. Does that fall into this category as well?

 

[DNCAgain]

Differences especially in the bass extension are still common and bass is super important for our listening pleasure.

Will the bass extension affect the overall sound of a speaker,or it simply brings a change in bass?

 

[SSS]

My opinion is that measurement is measurement and listening is listening and not measurement equals listening. Just to grab the frequency responses is not enough. There are many other parameters in the used speaker chassis, the case, etc. Between very expensive loudspeakers > 20 k€ I heard differences even in a treated room which was a hifi vendor room. Also the sound changed rapidly with slightly off axis. And of course the sound of loud bass differs between the models. Therefore both is necessary and after all the users taste decides.