What's Left In Speaker Design To Reduce Distortion/Increase Detail Retrieval?

[tuga]

Here is the links you cited, with all the relevant data, directivity and CSD. Your bluff was called.

If your coksuredness leads you to believe (yes) that you are in possession of enough information to accurately describe audibility of speaker linear and non-linear distortions as well as variations in frequency response and directivity you are delusional.

 

[MattHooper]

Meanwhile you audiophiles can't ever figure this out

So you aren’t an audiophile? On a site devoted to discussing audio gear?

Thanks for the red herring, though. ;-)

 

[Purité Audio]

A site devoted to the actual facts not anecdote.
Keith

 

[FrankW]

The Salon2 produces a massive directivity change above 8.5k and the CSD shows a lot of rubbish when compared to a Kef Reference 5
If your coksuredness leads you to believe (yes) that you are in possession of enough information to accurately describe audibility of speaker linear and non-linear distortions as well as variations in frequency response and directivity you are delusional.

Your hands are waving frantically and you are the claimant

 

[FrankW]

So you aren’t an audiophile? On a site devoted to discussing audio gear?

Thanks for the red herring, though. ;-)

Thanks for being unable to discern sighted vs unsighted "listening". And having no interest in audio science. On Audio Science Review

 

[Bjorn]

Doing a research on the audibility of poor directivity above 8.5 kHz would be more than daunting.

You would need and have take into account:
- Similar speakers where the only difference is directivity in the highs
- Make researchers with many different proximites to side walls and preferably also use different room widths
- Try many different acoustic setups/treatment, considering the audibility of discrete reflections will depend on how well others are attenuated

Not too mention using trained listeners with good hearing and testing with a great variety of music material.

I find it very odd that someone believes a simple study without taking into account all the variables believe this would be absolutely "science". And isn't the burden of proof in the hands of the one who claim something? Where are the studies that at least consider some of these variables that actually prove directivity issues above this frequency is not audible?

 

[MattHooper]

Thanks for being unable to discern sighted vs unsighted "listening". And having no interest in audio science. On Audio Science Review

You must be new here.

 

[MattHooper]

A site devoted to the actual facts not anecdote.
Keith

As I've made explicit numerous times: the truth or falsity of my experiences can be put aside and anyone can just discuss the question I have posed.

You know quite well that the anecdotes are presented with the proper caveats - as most of us do on this site. There are, thankfully, members here who manage to not be so dogmatic and can enjoy reports of ASR members getting out to hear different speakers, especially some discussed here, taking them for what they are. So long as they are not asserted as truths others need accept, which nobody does. Sighted listening can be subject to bias, but that does not entail it is useless or always inaccurate. (Otherwise, as I've pointed out, you tread in to incoherence).

That said, by all means, putting the issue of sighted preferences aside: Can you point to anything about the design of the Kii speakers, including any measurements, that suggest my description of the sound of the Kii 3s has been inaccurate?

 

[FrankW]

And isn't the burden of proof in the hands of the one who claim something?

The Salon2 produces a massive directivity change above 8.5k and the CSD shows a lot of rubbish when compared to a Kef Reference 5.

Absolutely agree, burden of proof always on claimant. We now know there was no there there.

 

[FrankW]

You must be new here.

So you have no interest in audio science, on an Audio Science forum. Interesting. Makes no sense to suggest Toole's book as a start to all your questions then.

 

[BoredErica]

Is it possible to have controlled directivity speakers (cardoid? like Kii 3) at far lower prices? Maybe in a decade we can have something $2k or less that when paired w/ a sub, provides very good FR while being unphased by SBIR?

Correct me if wrong. :^)

But ofc this is more a question of cost (and cost + aesthetics as if few speakers have it, the aesthetics are limited to whichever speakers have those features) rather than new things never done before.

 

[MattHooper]

So you have no interest in audio science, on an Audio Science forum. Interesting. Makes no sense to suggest Toole's book as a start to all your questions then.

You quoted a question of mine and veered off in to a red-herring, with ill-informed assumptions about my attitude toward science. I'm quite aware of Toole's work (we all are) and if you read more carefully you'd see I haven't made a single objective claim here that contradicts audio science. I'm betting I have done more to defend science (and blind testing), and it's relevance to audio to a great many resistant subjectivist audiophiles than you have. It's also possible - even likely I'd bet - that I've blind tested more of my equipment than you have.

I understand the hackles go up when some read subjective descriptions on this site. But maybe slow down and get some context, otherwise you gum things up by leaping to wrong assumptions.

 

[Tangband]

This question popped in to my mind simply from some recent experience listening to some speakers at another audiophile's place.

I currently listen to some smaller floor standing speakers with good quality drivers (Joseph Audio Perspective 2 Graphene) and I find there to be a gob-smacking sense of clarity
and detail in to recordings. Along the lines of "how could it get better than this?" (And I've heard lots of other speakers).

Then I go over to my Pal's place and listen to a pair of big ol' Estelon speakers, one of the newer "it" brands in high end audio circles. I forget which new model, but they retail for something like $65K. Now, most of us have had plenty of experiences showing us that money doesn't necessarily buy you any better sound in high end audio. But I have to say, even though the presentation ultimately wasn't to my liking as much as my own system, they just seemed to obviously dig out more sonic information in the recordings. So for instance drums on a track on my system would be well placed in spatial terms, and I can hear if the drums were placed in a reverb. But the Estelon speakers just seem to effortlessly carve out precisely where the drums are in the soundstage and the precise acoustics or added reverb around the drums...and exactly where that reverb "ends" is more vivid and obvious. Basically there is this constant sense of more sonic information, presenting more precision about what is in the recording.

Which had me wondering what accounted for these differences. Better drivers? The more heroic efforts that went in to removing the influence of the Estelon cabinets? The whole design?

Now, that's just accounting for why this question was on my mind. Anyone can simply ignore the above example (it's just my subjective impressions after all) but still get to the issue I'm wondering about:

What is left in terms of speaker design to achieve, in terms of lowering audible distortion and hence retrieving more neutral sonic information from recordings?

(I add "neutral" because of course one can always hype a speaker's high frequency response to increase perceived detail...that's not what I'm talking about).

Are we done? Or is there more to achieve in terms of materials and design (drivers, cabinets etc)? Is a very flat frequency response all there is (since resonances will purportedly show up in frequency response)? Or could we take a speaker that measures very even, yet some upgrade in driver material/design or even more reduction in cabinet resonances may yield even higher sonic performance, retrieving some subtle details that were obscured before?

Where can we go from here?

There are some demands to get there, except for good directivity and frequency response…:

1. Active dsp crossover with roomcorrection.
2. Amplifiers optimized with the optimal driving impedance for each driver- less distortion and less dynamic compression.
3. A 4-way speaker is suitable for a non compromising sound at home .
4. A dsp with feedback can cancel out some faults in the drivers.

Are we there yet ? Yes, but its expensive , and will cost less in the near future.

You can look at the new Genelec 8381A and Linn klimax 360 for good examples for future highend speakers with real performance leap.

 

[FrankW]

What is left in terms of speaker design to achieve, in terms of lowering audible distortion

What "audible distortion"?

and hence retrieving more neutral sonic information from recordings?

What "neutral sonic information"? How do you know what is "from recording" without transduction? Answer, you have no clue, but since you don't know about Circle of Confusion...self explained.

Are we done? Or is there more to achieve in terms of materials and design (drivers, cabinets etc)?

Based on what evidence any of that matters?

Is a very flat frequency response all there is (since resonances will purportedly show up in frequency response)?

Speakers a 3D radiators. "Flat frequency response" where? What distance and azimuth? They have thousands of "Frequency response". Probably missed Toole mentioning spin data...

Or could we take a speaker that measures very even, yet some upgrade in driver material/design or even more reduction in cabinet resonances may yield even higher sonic performance, retrieving some subtle details that were obscured before?

Based on what evidence? Oh wait, sighted "listening" revealing "new details" again...

I understand the hackles go up when some read subjective descriptions on this site. But maybe slow down and get some context, otherwise you gum things up by leaping to wrong assumptions.

Yeah that's it Matt.

 

[MattHooper]

What "audible distortion"?

Shirley, you can't be serious.

What do you think the goals are in engineering a good speaker driver and why?
What do you think the goals are in engineering speaker cabinets and why?
What do you think the goals are in various crossover designs, and why for instance they seek to suppress the influence of driver break up modes?
Why do you think most of the designs lauded by ASR members seek a generally flat on-axis frequency response (with evenly sloping off axis response)?
Etc etc.

If you are truly baffled about what "audible distortion" we could be discussing, look up: Linear Distortion. Harmonic Distortion. Intermodulation Distortion, and others.

What "neutral sonic information"? How do you know what is "from recording" without transduction? Answer, you have no clue, but since you don't know about Circle of Confusion...self explained.

Yes, that must be it. I'm a long time ASR member with thousands of posts and I've never heard of Toole's work or the Circle Of Confusion. Never discussed the issue ad nauseam either. You've got me there. Seriously...slow down...practice a bit of charity by actually engaging in conversation where you want to understand someone, rather than pose questions assuming "gotchas."

What I meant by "neutral sonic information" is sonic detail about a recording that comes from netural, low distortion reproduction, rather than from goosing up certain frequency ranges that either give the impression of "more detail" or that raise the amplitude and hence audibility of certain details that would otherwise be less noticed or discernible. I presume this phenomenon isn't totally foreign to you I hope? (BTW, I work in post production sound for film/TV - Sound Design - so I'm quite aware of how sound can be manipulated via EQ and many other ways to achieve various audible results).

Based on what evidence any of that matters?

Are you really suggesting the engineering in regards to speaker drivers, cabinet design, crossover design etc...doesn't matter? Or that...there is no evidence it matters?
Again...what in the world do you think audio engineers have been trying to achieve all these years when they set about designing speakers? Your desire to get to some
sort of pre-set "gotcha," smoking out rascally subjectivists, is leading you to some very strange posts.

Speakers a 3D radiators. "Flat frequency response" where? What distance and azimuth? They have thousands of "Frequency response". Probably missed Toole mentioning spin data...

Yes, never heard of spin data. On a site in which Toole participates and in which all speaker reviews recieve spin data. Again...you know me too well.

You are aware, right, that the work of Toole and others about why a "good" speaker design seeks generally flat on-axis response (again, not ignoring the influence of off axis behavior). You can put the goal in terms of maintaining accuracy to the signal coming out of your source/amp etc. Or you can simply put it in terms of preference - the blind testing research indicated that most people prefer the sound of a "neutral" speaker - generally flat on-axis sound, which is what our brains primarily use to interpret the sound of a source. In which case, there are all manner of ways a speaker can deviate from a flat frequency response, right? Audibly so. Right? This would be a "distortion" relative to the goal you are seeking to achieve, right? You can look up more about speaker design and figure out the rest I hope.

Based on what evidence? Oh wait, sighted "listening" revealing "new details" again...

There you go again. Why do you assume this?

If we want to be more sure of any such conclusions, the ideal way to evaluate would be blind listening, for instance similar to the type cited by Floyd Toole.

Please notice: Nowhere did I propose that the questions I asked were to be settled by sighted listening. And I have long defended the relevance of blind testing for any endevour in which we require stronger evidence than mere sighted anecdotes. This is why I've been explicit in this very thread (and many others) that nobody need take my anecdotes as settling anything at all on the matter, and that they can be ignored for the question at hand.

But you wouldn't know this because you are unaware of what I've argued for on this forum, and prefer to keep leaping to unfounded assumptions.

Yeah that's it Matt.

Yes, you were wrong about all your disparaging suggestions. Wrong that I didn't know or didn't care about the relevance of science and blind testing to audio gear. Wrong about my being unfamiliar with Floyd Toole's work and it's implications. Wrong that I was assuming sighted listening like mine settles any such questions. And not a single question you posed results in the "gotcha" you seem to have assumed. (And...in terms of actual personal experience: how many blind tests have you participated in? Just curious).

 

[Axo1989]

Not really worth clicking "show ignored content" to follow that discussion, but good on y'all for trying.

 

[MattHooper]

There are some demands to get there, except for good directivity and frequency response…:

1. Active dsp crossover with roomcorrection.
2. Amplifiers optimized with the optimal driving impedance for each driver- less distortion and less dynamic compression.
3. A 4-way speaker is suitable for a non compromising sound at home .
4. A dsp with feedback can cancel out some faults in the drivers.

Are we there yet ? Yes, but its expensive , and will cost less in the near future.

You can look at the new Genelec 8381A and Linn klimax 360 for good examples for future highend speakers with real performance leap.

Good post, thanks.

Regarding #2 and #4: One of the knocks usually against, say, a large 2-way speaker attempting to get down to 30Hz or below, is not only the problem of driver matching to the tweeter (among them increasing beaming at the top of the woofer's passband mismatching with the tweeter's dispersion) but also that such a design is more likely to suffer higher levels of distortion when playing bass heavy content than, for instance a good 3 way design. It's also *one* of the reasons subwoofer fans cite for crossing over even floor standing speakers to subwoofers: takes some of the load off the speaker's woofers so in principle there will be less distortion from the floorstanding speaker when playing demanding content.

I'm wondering how likely this is to actually be audible. I've seen anecdotal reports from subwoofer fans of less apparent "strain" to the sound at high volumes once mains are crossed to a sub, but how much distortion does it take until this becomes an audible benefit? I had my Thiel speakers (spec'd down to 35Hz) crossed over (varying between 40 to 100 Hz, usually 80Hz) and don't remember a reduction in audible distortion/strain to the sound. But then, I don't listen really loud.

What has anyone here measured in that regard?

(What I'm getting at is that if one is expecting on theory less distortion from the main speaker's drivers when using a sub, one might be inclined to "hear" less "strain" or distortion in a biased way. What therefore do we have in terms of measurements and how high does the distortion have to get to be audible, such that crossing to a sub gains audible benefits in that sense).

 

[MattHooper]

Not really worth clicking "show ignored content" to follow that discussion, but good on y'all for trying.

Not worth it.

I've never used it before, but my ignore button finger is gettin' itchy...

 

[tuga]

Absolutely agree, burden of proof always on claimant. We now know there was no there there.

Why are fundamentalists so righteous...and wrong?

 

[tuga]

As I've made explicit numerous times: the truth or falsity of my experiences can be put aside and anyone can just discuss the question I have posed.

You know quite well that the anecdotes are presented with the proper caveats - as most of us do on this site. There are, thankfully, members here who manage to not be so dogmatic and can enjoy reports of ASR members getting out to hear different speakers, especially some discussed here, taking them for what they are. So long as they are not asserted as truths others need accept, which nobody does. Sighted listening can be subject to bias, but that does not entail it is useless or always inaccurate. (Otherwise, as I've pointed out, you tread in to incoherence).

That said, by all means, putting the issue of sighted preferences aside: Can you point to anything about the design of the Kii speakers, including any measurements, that suggest my description of the sound of the Kii 3s has been inaccurate?

The best Purité Audio Keith is the Ignored Purité Audio Keith.