What role should listening play in speaker design

[PristineSound]

I don’t know about Chesky but Borresen employs measurements along with their listening and certainly so does Danny (his constantly showing measurements on his channel for his speaker corrections).

So realistically, it really comes down to discussing speakers that are designed not only with measurements but also via informal (non-blind) listening playing a part.

And if you take a peek just a few posts up, you’ll see I presented an example in Joseph Audio speakers. (and that’s just a specific example from countless other possible examples)

Cheers

Frankly, I believe Chesky does do measurements, the kids interned at 3D3A lab at Princeton, so I am pretty sure he knows something about measurements and anechoic chambers, etc., does he know about the research conducted by Toole, I would assume so, because he would be pretty dumb to send a pair to Erin for review if he doesn't understand some basic concept of CEA2034. I think the issue with Chesky may be that he simply doesn't have access to good measurement tools, understanding of how to correctly measure and/or don't have much experience in designing speakers; if that is the case, this is why you shouldn't trust your damn ears for industrial precision measurements.

 

[a4eaudio]

.... and certainly so does Danny (his constantly showing measurements on his channel for his speaker corrections).

That makes sense, it might have been distortion measurements that he "didn't believe in".

 

[a4eaudio]

Frankly, I believe Chesky does do measurements, the kids interned at 3D3A lab at Princeton, so I am pretty sure he knows something about measurements and anechoic chambers, etc., does he know about the research conducted by Toole, I would assume so, because he would be pretty dumb to send a pair to Erin for review if he doesn't understand some basic concept of CEA2034. I think the issue with Chesky may be that he simply doesn't have access to good measurement tools, understanding of how to correctly measure and/or don't have much experience in designing speakers; if that is the case, this is why you shouldn't trust your damn ears for industrial precision measurements.

All he needs is a $60 Dayton EMM-6 mic, a 2-in, 2-out USB interface ($120-$200) and REW ($Free).
Chesky did not send the speaker to Erin, a consumer purchased the speakers and sent them to Erin.

 

[RexrothPigeon]

Yes, the problem is that ruler flat on-axis doesn't tell us everything we need to know.

Good thing that isn't the only measurement taken here, come on be serious

There are plenty of variables for taste, including directivity, bass extension, SPL, etc., but ruler-flat on-axis (or very close) is a bare minimum for a well designed speaker in 2026, when DSP and room correction are accessible to all.

 

[PristineSound]

Chesky did not send the speaker to Erin, a consumer purchased the speakers and sent them to Erin.

ooh, well, that says it all.

But on the other hand, at CAF this year, his room was packed with a bunch of people who thought it sounded great. It has deep bass that didn't have too much audible distortion, so people were impressed by that. I wasn't one bit impressed.

 

[amirm]

Going back to the title of the thread... What speakers are KNOWN to be voiced by ear and not using measurements as an important input? I think this is true for Borresen, Chesky and Danny Ritchie but I gladly admit that I don't pay attention to those threads, so I may be mistaken.

I think you can add a ton of high-end speakers to the list. Wilson would be one that comes to mind right away.

 

[amirm]

Just so that it is clear, what bothers me about AJ's comment is that he is basically saying, "if you find measurement errors, it is no big deal because I decided on something better." Same thing I hear from electronics makers who don't properly optimize their design using measurements. We are never given evidence, e.g. controlled listening tests, that the compromise he has made by ear was the best. As someone said, you are supposed to believe the Guru and that is that.

 

[Everett T]

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 measurements. He also throws dart at the accuracy of such measurements. Here is one of countless interviews with the same theme:

Are speakers really better when the designer tweaks them by listening?

Thanks for starting the discussion! He really is talking down to everyone, and is extremely cocky.

Measuring after each change to the crossover or box till the speaker is right and it should be close to the design goal. If using automated measurement system, the process is much more accurate. Voicing by ear before the measuments and after seems backwards.

It doesn't require much time after the measurements goal is achieved. I wouldn't need 10 songs to make sure the speaker is right and was being generous to Andrew. These are genres I use to finalize the speakers: female vocals, trios, soloist, and dynamic classical or opera.

 

[PristineSound]

Just so that it is clear, what bothers me about AJ's comment is that he is basically saying, "if you find measurement errors, it is no big deal because I decided on something better." Same thing I hear from electronics makers who don't properly optimize their design using measurements. We are never given evidence, e.g. controlled listening tests, that the compromise he has made by ear was the best. As someone said, you are supposed to believe the Guru and that is that.

Completely agree with everything you said, but I do want to point out one thing though.

We all have been spoiled by Dr. Toole and Dr. Olive, because I recall reading somewhere that as part of their employment agreement with Harman, they were allowed to publish their scientific research. Not all companies are willing to publish research they have, as a matter of business priorities.

This statement does imply me saying that AJ or others gurus doing "voicing" have solid scientific evidence from a controlled experiment. I'm just saying that we have been spoiled.

 

[SigbergAudio]

Good thing that isn't the only measurement taken here, come on be serious

There are plenty of variables for taste, including directivity, bass extension, SPL, etc., but ruler-flat on-axis (or very close) is a bare minimum for a well designed speaker in 2026, when DSP and room correction are accessible to all.

I've explained several times why that isn't necessarily the case, but feel free to hold on to your beliefs.

 

[RexrothPigeon]

I've explained several times why that isn't necessarily the case, but feel free to hold on to your beliefs.

your explanations have been stuff like "I like other speakers better than the kii" and "there are things measurements don't show," so yeah, I remain unconvinced

 

[SigbergAudio]

Just so that it is clear, what bothers me about AJ's comment is that he is basically saying, "if you find measurement errors, it is no big deal because I decided on something better." Same thing I hear from electronics makers who don't properly optimize their design using measurements. We are never given evidence, e.g. controlled listening tests, that the compromise he has made by ear was the best. As someone said, you are supposed to believe the Guru and that is that.

I never heard him say that. I heard him say that it is difficult to trust the measurements, not because there's something inherently wrong with measurements, but because it's hard to take accurate measurements and there is no reference to use to check if you did it right.

If we for the sake of discussion assume the Klippel is perfect, that still doesn't make it easy, as that tool costs 100,000USD.

 

[SigbergAudio]

your explanations have been stuff like "I like other speakers better than the kii" and "there are things measurements don't show," so yeah, I remain unconvinced

No, neither of those were arguments against ruler flat on-axis response. You are reading every counter argument with obvious bias that you already know that you are right, if you are reading them at all. Curiosity is always useful if you want to learn something new. If you don't want to learn something new, that is of course perfectly fine.

 

[RexrothPigeon]

there is no reference to use to check if you did it right

If for the sake of discussion we assume that measurements are hard to verify and can't always be trusted, well, the same is doubly or triply true for one person's ears, especially if that person is a septuagenarian.

 

[SigbergAudio]

It doesn't require much time after the measurements goal is achieved. I wouldn't need 10 songs to make sure the speaker is right and was being generous to Andrew. These are genres I use to finalize the speakers: female vocals, trios, soloist, and dynamic classical or opera.

Which speakers have you designed?

 

[SigbergAudio]

If for the sake of discussion we assume that measurements are hard to verify and can't always be trusted, well, the same is doubly or triply true for one person's ears, especially if that person is a septuagenarian.

Strawman.


For the benefit of other non-English speakers like myself:

Septuagenarian

1. a person who is between 70 and 79 years old.
   
   
   
   
   

 

[RexrothPigeon]

For the benefit of other non-English speakers like myself:

Septuagenarian

1. a person who is between 70 and 79 years old.

I believe Andrew Jones is in fact in that age range, but definitely correct me if I'm wrong!

 

[SigbergAudio]

I believe Andrew Jones is in fact in that age range, but definitely correct me if I'm wrong!

I have no idea, I had just never heard the word before.

 

[Sokel]

We have to discriminate the task of each speaker.

Given for example a speaker that can be ruler -flat anechoically and SPL adequate, and low distortion across the spectrum enough for 3 meters distance,
it will be just fine at this distance but can be thin and strident at 1 meter, as its FR won't decline enough for it.

Measurements taken lightly may not account for that that's why companies has to state the intended use, as Genelec does for example.

 

[a4eaudio]

If for the sake of discussion we assume that measurements are hard to verify and can't always be trusted, well, the same is doubly or triply true for one person's ears, ...

"Two ears and a brain respond very differently to a complex sound field — and are much more analytical — than an omni-directional mic and analyzer. "