The Insanity of Marketing Disguised as Science in Loudspeakers

[zhimbo]

How do you make a Stradivarius sound like a Stradivarius, we don't even know what is special that makes this instrument sound like this.... i

Violinists can’t tell the difference between Stradivarius violins and new ones

Antique Italian violins, such as those crafted by Antonio Stradivari or Giuseppe Guarneri “del Gesu”, can fetch millions of dollars. Many violinists truly believe that these instruments are better than newly made violins, and several scientists have tried to work out why. Some suspected at the unusually dense wood, harvested from Alpine spruces that grew during an Ice Age. Others pointed the finger at the varnish, or the chemicals that Stradivari used to treat the wood.

But Claudia Fritz (a scientist who studies instrument acoustics) and Joseph Curtin (a violin-maker) may have discovered the real secret to a Stradivarius’s sound: nothing at all.

There's been quite a few such tests of Strads, and so far the story is pretty darn consistent: People can easily pick out the "Strad sound"...when they already know they're listening to a Strad. There are, of course, plenty of objections to all the tests, but I'm personally shocked at the consistency of the findings again and again.

 

[Nuance AH]

Dunlavy Closes Doors | Stereophile.com

And an interesting quote from one of the speaker reviews:

"DAL firmly believes that a full set of credible measurements, made by qualified engineering staff using state-of-the-art equipment and facilities, can reliably predict the potential of a loudspeaker to accurately reproduce the complex sounds of music."—Dunlavy Audio Labs

Dunlavy Audio Laboratories SC-IV loudspeaker | Stereophile.com

And a review that has off-axis measurements (that don't look great for that matter):

Dunlavy Audio Labs SC-IV/A loudspeaker | Stereophile.com

I noticed Dunlavy basically tells John his measurement technique was wrong when measuring these speakers. I've seen Richard Vandersteen do the same, which begs the question: Do time and phase accurate speakers need to be measured differently in-room?

 

[AcuDefTechGuy]

I noticed Dunlavy basically tells John his measurement technique was wrong when measuring these speakers. I've seen Richard Vandersteen do the same, which begs the question: Do time and phase accurate speakers need to be measured differently in-room?

Well, the time coherent measurements look fabulous. So that part of the measurement was correct. The spectral decay measurements look good, so they are correct.

Only the off-axis measurements (& some on-axis - Dunlavy IV was like +/-4dB on-axis) look bad, so they were incorrectly measured.

 

[GranteedEV]

I noticed Dunlavy basically tells John his measurement technique was wrong when measuring these speakers. I've seen Richard Vandersteen do the same, which begs the question: Do time and phase accurate speakers need to be measured differently in-room?

I think it has to do with all those drivers in these large multi way speakers. They take a given distance to sum as the designer intended. That not only applies to the microphone, but to your ears.


Fine, let's say the speaker is +/- 1db on axis. The real issue... the off-axis response still doesn't track it. Those stereophile polars are normalized to the axial response. So even if the axial response were perfect, the off-axis response doesn't match it sufficiently.

 

[cschang]

I noticed Dunlavy basically tells John his measurement technique was wrong when measuring these speakers. I've seen Richard Vandersteen do the same, which begs the question: Do time and phase accurate speakers need to be measured differently in-room?

I thought the Dunlavys were measured anechoically.

I didn't thoroughly read the articles though.

 

[Nuance AH]

I thought the Dunlavys were measured anechoically.

I didn't thoroughly read the articles though.

That is how I read into it as well, but in the manufacturer's comment section Dunlavy states this:

"We are equally indebted to John Atkinson for his measurements of the SC-IV/A, but only wish that Stereophile possessed a more extensive array of laboratory test equipment and a large anechoic chamber. Such a facility is necessary because loudspeakers having the physical size of the SC-IV/A must be measured "anechoically" at the normal listening distance of 10' if the measurements are to convey any relevance as to how accurately they will reproduce complex musical sounds and transients in a typical listening room."

I'm confused... With that said, the impulse and step response look quite good, so...

I find it interesting that this is one of the only reviews I've seen in which John knocks a speaker. He mentions cabinet resonance (really???) and a push in the upper midrange and treble. I guess Dunlavy didn't pay him enough advertising $$$

Edit: correction, John says it's a well engineered speaker, but later on goes on to knock it in the follow up. Waa waa whaaaat? Dude needs to make up his mind.

 

[haraldo]

So their off-axis response is terrible?

No... but the use of quite many drivers means that time coherence is not correct everywhere and it's about simple geometry that they will be phase coherent straight in front of the drivers at... say 10' .... if you go to the side the distance to the drivers change slightly so you lose the perfect time coherence, and as such you also lose focus a bit....

I do have the same thing with my Duntech PCL-15's, to some extent... even if they're only two way with two drivers.... but in spite of this I would never exchange them for speakers with higher order x-overs (but that's just my humble opinion)

That's why Thiel uses coaxial drivers, in order to get around this exact issue, perhaps these speakers because of this doesn't have these same issues....

 

[haraldo]

I noticed Dunlavy basically tells John his measurement technique was wrong when measuring these speakers. I've seen Richard Vandersteen do the same, which begs the question: Do time and phase accurate speakers need to be measured differently in-room?

Yes..... Dunlavy's point was that you have to measure them anechoically, because the ear/brain can sort out the reflections if they're delayed somewhat so that what you interpret is what comes from the speakers.... I'm not an expert in this, many others around here in this thread knows a lot more probably, but something about this is his point.

Actually Dunlavy relied mainly on step responses when measuring speakers, because using a step response everything must be in order or it will look bad.... you don't get a good step response if speakers are non linear, and if there are phase aberrations... the same.... so by one single measurement you can check mostly all aspects of the speaker (well not polar response.....)

 

[GranteedEV]

so by one single measurement you can check mostly all aspects of the speaker (well not polar response.....)

... or pistonic behavior, or power compression.


Three rather important things for high end speakers.

 

[Irvrobinson]

... or pistonic behavior, or power compression.


Three rather important things for high end speakers.

Though it has been ten years or so, I still remember the Dunlavy V auditions. The bass was so "to die for" I drove the dealer crazy by coming back about five times to listen. I suppose the midrange was too. There was something about the highs that were never quite convincing to me though. And the price at the time ($20K+... I don't remember) was really beyond my means, so I was looking for perfection at that price. Maybe it was just flawed ears and not flawed speakers. I was always a measurement fan, and I remember chasing them down due to the Stereophile review of the VI. If you look at JA's frequency response measurements they look almost unbelievable. I wonder what I would think of them now?

I can see why heraldo wants a pair. (Incidentally, there's a pair for sale on Audiogon for $7900. Tempted?)

 

[haraldo]

... or pistonic behavior, or power compression.


Three rather important things for high end speakers.

Yes, are you just trying to be difficult or is it in your nature.....
the step response replaces a bunchload of measurements, but not all of them

EDIT: (Removed some stuff)

 

[Irvrobinson]

Yes, are you just trying to be difficult or is it in your nature.....
the step response replaces a bunchload of measurements, but not all of them

Please reread my above post !

Sometimes he's like that, but overall GranteedEV is a key asset to this forum.

 

[AcuDefTechGuy]

We really need AJinFLA back here. Can we pretty please lift the ban?

I like learning from AJinFLA & GranteedEV.

 

[haraldo]

Sometimes he's like that, but overall GranteedEV is a key asset to this forum.

Yes agree 200%

 

[jinjuku]

We really need AJinFLA back here. Can we pretty please lift the ban?

I like learning from AJinFLA & GranteedEV.

I'm there with you. Like everyone AJ could have his moments. But when it was all said and done he backed up his points quite well. I never understood his ban.

 

[MinusTheBear]

Yes, are you just trying to be difficult or is it in your nature.....
the step response replaces a bunchload of measurements, but not all of them

EDIT: (Removed some stuff)

I don't quite understand this. So what would you be willing to replace/sacrifice aka trade-off in order to get a great step response?

 

[haraldo]

I don't quite understand this. So what would you be willing to replace/sacrifice aka trade-off in order to get a great step response?

I'm not willing to trade off anything.... I want utter perfection!
I know it doesn't exist, but we should look for this.....

In my view it's not an optimal trade-off to take away phase/time coherence because it's too difficult to deal with this....
If I'm rude... and I am sometimes, then I believe many loudspeaker manufacturers throws the phase coherence out the window because it's too difficult for them to deal with....

One problem of course with mostly all products using shallow slope crossovers is the load on the drivers so dynamics may suffer....

 

[AcuDefTechGuy]

I would love to buy a speaker that has perfect everything.

But none do.

In our example of the Dunlavy, it seems to excel in certain areas (time coherent?) while lacking in other areas (off-axis, polar response, power response?).

I guess some people prefer time coherence over off-axis, polar & power response?

The other question is, what is acceptable or inaudible or insignificant time/phase coherence?

Are we talking something like THD of 0.0000001% vs. 1% here?

 

[haraldo]

I would love to buy a speaker that has perfect everything.

But none do.

In our example of the Dunlavy, it seems to excel in certain areas (time coherent?) while lacking in other areas (off-axis, polar response, power response?).

I guess some people prefer time coherence over off-axis, polar & power response?

The other question is, what is acceptable or inaudible or insignificant time/phase coherence?

Are we talking something like THD of 0.0000001% vs. 1% here?

OK, there must be some sacrifices and if you want phase/time coherence you probably have to at least sacrifice on maximum SPL, I'm not so sure about off axis, polar response.... you probably know much better than me....

With thiel's and VAF's that do have coaxial tweeter midrange I guess there is better chance to keep the off axis response, with the huge array's like the big dunlavy's and big duntech's you probably will always have challenges with room placement and also off axis response due to the large set of drivers.....

In general though I believe that biggest sacrifice that you will have to accept is in max spl with these speakers... if this is acceptable or not is a question....

It's also been mentioned about some of these bigger dunlavy's and duntech's that room treatment is of more importance than with many other speakers, so this makes for a big challenge at home....

First you need huge speakers.... and then comes room treatments.... LOL
You need a good and nice better half... to have these big speakers.....

Can you not make a line array with a large amount of full range drivers and a line of tweeters and use something like a Behringer Ultracurve or a DEQX to deal with.... in order to have phase/time perfection, good off axis and polar response and almost limitless dynamics, I just ask?

Perhaps I'm going off thread now, please correct me if I am.....

 

[Hocky]

Technically, you can. scroll back a page or two where Irv robinson linked a white paper from pi speakers. Controlled, but-highish (but not beaming) directivity + Crossfiring will get you very close to this exact scenario of hearing both speakers at equal levels.

Moving away from center means more distance from one speaker, less distance to the other. Than means that you can easily go from 0/0 to +6db/-6db with just a few feet of movement. It doesn't matter what your speaker does, you can't change that.