Worst measuring loudspeaker?

[Purité Audio]

If you haven't listened to a pair of Klipschorns perhaps you should? I think you will find it an enlightening experience.

Dave.

I have heard them in customers houses on two occasions, I have no particular wish to hear another pair.
Keith

 

[Juhazi]

Thanks J, are there any recognised audible levels of distortion in loudspeakers?
Are there just too many variables to achieve a consensus?
Keith

No widely recognized. Too many buts and ifs. 30% is audible to golden ears if it is wide enough range - typically bass but there we are not so sensitive.

 

[Hayabusa]

No widely recognized. Too many buts and ifs. 30% is audible to golden ears if it is wide enough range - typically bass but there we are not so sensitive.

30? you mean 3.0?

REW has a generator where you can add harmonic distortion to a sine wave. Good exercise to find out what distortion you can hear.

 

[peanuts]

Zu audio agreed. one of the worst speakers iv ever listened to. dont remember model, but they all share the same 10" crossover-less wideband, cant possibly work in a million years. horrid high-end distortion, loads of beaming and an on-axis response like a rollercoaster.
yet Sean is still sticking to his guns, that a crossover ruins dynamics or whatever nonsense.

 

[Juhazi]

30? you mean 3.0? REW has a generator where you can add harmonic distortion to a sine wave. Good exercise to find out what distortion you can hear.

Sine wave is too easy to detect distortion. In context of loudspeaker measurements, we typically see most distortion in the low end (bass) and it is nonlinear by nature. Ear is not very sensitive to bass distortion, that's why I said 30%

http://www.aes.org/e-lib/browse.cfm?elib=19329
https://www.audioholics.com/loudspeaker-design/audibility-of-distortion-at-bass
Nonlinear distortion, on the other hand, is dependent on the level and frequency of the input signal and can often create many output frequencies based on a single input frequency. Nonlinear distortion is much more difficult to predict and assess, because you have to know what the input signal is like in order to understand its relation to the output, whereas in linear distortion, the distorting mechanism will treat every incoming signal the same way.

Closed box speaker with SEAS T18REX

 

[Ron Texas]

Towers
https://www.stereophile.com/content/volti-audio-rival-loudspeaker-measurements
https://www.stereophile.com/content/auditorium-23-hommage-cinema-loudspeaker
https://www.stereophile.com/content/bowers-wilkins-683-s2-loudspeaker-measurements
https://www.stereophile.com/content...and-symphony-edition-loudspeaker-measurements
https://www.stereophile.com/content/nola-metro-grand-reference-gold-loudspeaker-measurements
https://www.stereophile.com/content/voxativ-ampeggio-loudspeaker-measurements

Standmount
https://www.stereophile.com/content/bowers-wilkins-805-d3-loudspeaker-measurements
https://www.stereophile.com/content/wilson-benesch-square-one-series-2-loudspeaker-measurements

The two B&W speakers received great reviews from Stereophile. The 683 S2 was rated B full range and the 805 D3 rated A Restricted LF. I did not check the others. That doesn't make you wrong. It's just evidence of a different point of view. Professional reviewer Diego Estan owns a pair of the new 705 stand mounts and sometimes complains of harshness. Where there's smoke there's fire. I owned the 683 S2's before I had a calibrated mike. They were not growing on me. A flood destroyed them before they were a year old.

 

[Kal Rubinson]

A flood destroyed them before they were a year old.

Covered by insurance?

 

[suttondesign]

Old Bozaks. People who love them stick with 'em. That said, they sounded somewhat better than any big Klipsch I ever heard.

 

[Ron Texas]

Covered by insurance?

Losses were in excess of contents coverage. Whether this loss was covered or not is a matter of interpretation. The house was beyond economic repair and was eventually torn down. I live in a different part of town now.

 

[napilopez]

Neumann KH80 because no matter how good it measures we're still going to complain about it

In all seriousness, I've been lucky to not have to measure any truly terrible speakers, but the B&W formation duo is definitely... puzzling. The on-axis is actually decent. But the directivity characteristics... well:

And now vertical. Note these dips are only out to 15 degrees:

I wish I'd been able to calculate a full spin. Who knows, maybe it balances out with more data. But it's not looking good.

But then, they do have a lot of bass, and don't sound too bad on axis. Soundstage is finnicky though.

 

[mhardy6647]

If you haven't listened to a pair of Klipschorns perhaps you should? I think you will find it an enlightening experience.

Dave.

Actually, to my ears, and FWIW -- the K-horns sound pretty good compared to the other "Heritage" Klipsch models.
I believe that the Geneva Convention has regulated the use of the LaScala and Heresy during periods of conflict.

https://www.audioasylum.com/cgi/vt.mpl?f=hug&m=177391
(ahem)

 

[thewas]

In all seriousness, I've been lucky to not have to measure any truly terrible speakers, but the B&W formation duo is definitely... puzzling. The on-axis is actually decent. But the directivity characteristics... well

B&W sounds them like that (with a "BBC dip" in the sound power) because its currently their brand sound, it's not that they couldn't do better, just look at the measurements of their models till the early 2000s:

(entry series model DM603S3 from https://www.stereophile.com/content/bw-dm603-s3-loudspeaker-measurements)

Suddenly they changed their strategy (low order and high frequency mid/tweeter crossover which result such a behaviour) but the sales didn't punish them, rather the opposite, it's quite a sweet euphonic voicing which makes anything sound not tiring and impressive, a bit like the modern TVs in showroom mode or the rental car radios with bass and treble fully boosted as Toole writes.

I wish I'd been able to calculate a full spin. Who knows, maybe it balances out with more data.

It actually did on the current B&W 606 which I had for some months and except the "BBC dip" needed less EQ above the modal region to meet a Harman target on my listening position compared to my KEF LS50:

(MMM measurement of the B&W 606 at my listening position without any EQ)

 

[napilopez]

B&W sounds them like that (with a large "BBC dip" in the sound power) because its currently their brand sound, it's not that they couldn't do better, just look at the measurements of their models till the early 2000s:

View attachment 49592View attachment 49593
(entry series model DM603S3 from https://www.stereophile.com/content/bw-dm603-s3-loudspeaker-measurements)

Suddenly they changed their strategy (low order and high frequency mid/tweeter crossover which result such a behaviour) but the sales didn't punish them, rather the opposite, it's quite a sweet euphonic voicing which makes anything sound not tiring and impressive, a bit like the modern TVs in showroom mode or the rental car radios with bass and treble fully boosted as Toole writes.

Yep, I remember seeing some of their old measurements and seeing how much better is sounded. You can see some of how the sound ends up balancing out in this in-room averaged measurements (Ignore the nulls I'd normally fill in with a sub). In a way it's actually better than I expected, but it's nonetheless a rather prominent dip around 2K to 9K.

 

[andreasmaaan]

Neumann KH80 because no matter how good it measures we're still going to complain about it

In all seriousness, I've been lucky to not have to measure any truly terrible speakers, but the B&W formation duo is definitely... puzzling. The on-axis is actually decent. But the directivity characteristics... well:

And now vertical. Note these dips are only out to 15 degrees:

I wish I'd been able to calculate a full spin. Who knows, maybe it balances out with more data. But it's not looking good.

But then, they do have a lot of bass, and don't sound too bad on axis. Soundstage is finnicky though.

The vertical polar response is as expected given the high xo point and the distance between the tweeter and woofer. The higher the inverse ratio of C2C distance to XO frequency, the closer to the central axis the cancellation null will occur.

The horizontal polar response is also explicable: the woofer, having a larger cone, begins to beam off-axis as frequency increases. The speaker then becomes less directional above c. 4kHz as the tweeter's output begins to dominate.

What makes less sense is why this behaviour doesn't seem to be visible in the vertical polar response. How far out does each measurement go?

 

[napilopez]

The vertical polar response is as expected given the high xo point and the distance between the tweeter and woofer. The higher the inverse ratio of C2C distance to XO frequency, the closer to the central axis the cancellation null will occur.

The horizontal polar response is also explicable: the woofer, having a larger cone, begins to beam off-axis as frequency increases. The speaker then becomes less directional above c. 4kHz as the tweeter's output begins to dominate.

What makes less sense is why this behaviour doesn't seem to be visible in the vertical polar response. How far out does each measurement go?

Sure, I just mean to say most modern manufactures at least try to prevent such a massive crossover dip. I've not measured another speaker that exhibits on quite so big. But I guess it's kind of interesting that they try to put the BBC dip in the power response rather than the on axis one.

For the vertical measurements, I noted those measurements are in 5-degree increments; 5/10/15 degrees off axis. So I assume they become more controlled further out, but at the time I wasn't doing full 360-degree measurements.

 

[Poseidons Voice]

Here’s more to see:
https://www.stereophile.com/content/zu-essence-loudspeaker-measurements

https://www.stereophile.com/content/reference-mm-de-capo-i-loudspeaker-measurements

https://www.stereophile.com/content/coincident-speaker-technology-troubador-loudspeaker-measurements

https://www.stereophile.com/content/shahinian-diapason-loudspeaker-system-measurements

Best,
Anand.

 

[Blumlein 88]

Bozak Grand

Stereophile measurements. The Bozak is maybe the worst off axis vs the Zu and K-horn.

Bozak lateral response.

Zu Essence for comparison.

K-horn comparison.

BTW, how do you evaluate the K-horn with the Klippel as the corners of the room are supposed to be an important part of the speaker. Anechoic isn't a picture of how they'll sound in a good room with large solid corners? Also NOTE:the AK6 has an enclosed cabinet and isn't quite the classic K-horn where the room finishes the horn exit.

Seems this speaker is the source of the old hifi terms. Woof, squawker, and tweet.

 

[andreasmaaan]

Sure, I just mean to say most modern manufactures at least try to prevent such a massive crossover dip. I've not measured another speaker that exhibits on quite so big. But I guess it's kind of interesting that they try to put the BBC dip in the power response rather than the on axis one.

For the vertical measurements, I noted those measurements are in 5-degree increments; 5/10/15 degrees off axis. So I assume they become more controlled further out, but at the time I wasn't doing full 360-degree measurements.

I see. Sorry, had misunderstood you. Yeh, it hasn't been fashionable to cross mid to tweeter that high in frequency for many years. Pre-1970s, around 5kHz was a more common mid-to-tweeter XO point (based on my limited knowledge of these old designs, I must say). So these B&Ws in that sense share more with much older speakers.

I'm not sure they've tried to put the BBC dip in the power response though. I think they've simply tried to cross a large-ish midwoofer with a dome tweeter at 5kHz to create a flat(ish) axial response. The dip in the polar response below the XO point is simply a result of the directivity mismatch between the two drivers. If they were interested in giving this speaker a BBC-style dip, they would have reduced the tweeter's axial output in the 5-8kHz region, i.e. at and just above the XO point. Looking at the measurements, I'm not sure they're even thinking about the speaker's power response too much in this design...

 

[Frank Dernie]

I went to a talk by a Linn speaker engineer who said they target a "smiley" FR curve because that is what people like best.
I presume this was based on some research, and certainly there are plenty of modern speakers with this sort of target (though obviously not those Harman disciples).

 

[thewas]

I went to a talk by a Linn speaker engineer who said they target a "smiley" FR curve because that is what people like best.
I presume this was based on some research, and certainly there are plenty of modern speakers with this sort of target (though obviously not those Harman disciples).

Even not few from the Harman group itself , just few examples that 30 seconds of googling gave
https://www.soundstagenetwork.com/measurements/jbl_l890/
https://www.stereophile.com/content/jbl-synthesis-1400-array-bg-loudspeaker-measurements
https://www.stereophile.com/content/jbl-studio-l880-loudspeaker-measurements