KEF Q100 Speaker Review

[QMuse]

I always say speaker business is 80% about marketing and KEF has mastered that art.
- amirm

Only JBL 305II has shined so far I guess. The Genelics too.

From what I can tell when comparing the score table with Klippel's spinorama graphs the only true liebling of scoring system is Harbeth.

 

[edechamps]

Everything you said is true and it does to some extent affect precision of THD measurement. Unfortunately all that in the same manner also affect frequency response measurements but I don't remember anybody complaining. In fact there have been many comparisons of various FRs made by different sources and in totally different venues.

FR measurements are more standardized than non-linear distortion measurements. For example, unless otherwise noted, one naturally expects a speaker FR measurement to be done in far-field anechoic free space conditions (simulated or actual). Speaker FR measurements are not expected to be level-dependent unless there is something seriously wrong with the speaker or it is driven into compression (which should not be the case unless that is explicitly stated as part of the measurement). Background noise is not a big issue with typical FR measurement techniques.

In the end, speaker FR measurements are sufficiently standardized that multiple sources with different protocols can often be compared directly, as is often done in these review threads (with S&R, etc.). Different protocols might produce results with varying accuracy (e.g. anechoic vs. quasi-anechoic vs. NFS), but on average one can expect them to converge on the same numbers, and in practice they do, as has been shown many times.

In contrast, non-linear distortion doesn't follow any widely agreed-upon standard (aside from the definition of the THD formula itself, which is a small part). You might get wildly different results depending on drive level, test signal used (frequency, shape), background noise, acoustical environment, and perhaps other factors like short-term vs. long-term distortion. Protocols are all over the place and it's rare to find two protocols that are sufficiently similar that the results can be compared directly. This is not surprising because there is still a lot we don't know about how to make non-linear distortion measurements that actually correlate well with human perception, so this area is still very much in the "experiment"/"trial and error" stage as opposed to the "standardization" stage.

 

[b1daly]

Amir’s comment on a kind of buzzing distortion mirrors what I heard when I bought a pair of KEF Q150s on a whim. Overall that speaker struck me as having a decent frequency response but that distortion rendered them unlistenable. I guess I would describe it as a kind of harshness. It was unfamiliar to me, and I wondered if it was an attribute of the coaxial driver?

 

[hardisj]

Also @hardisj did you measure xmax for this driver?

I believe so. But, again, I can't even access my own site from work so there's that... lol

It should be listed under the "woofer testing" portion of the review.

 

[QMuse]

FR measurements are more standardized than non-linear distortion measurements. For example, unless otherwise noted, one naturally expects a speaker FR measurement to be done in far-field anechoic free space conditions (simulated or actual). Speaker FR measurements are not expected to be level-dependent unless there is something seriously wrong with the speaker or it is driven into compression (which should not be the case unless that is explicitly stated as part of the measurement). Background noise is not a big issue with typical FR measurement techniques.

In the end, speaker FR measurements are sufficiently standardized that multiple sources with different protocols can often be compared directly, as is often done in these review threads (with S&R, etc.). Different protocols might produce results with varying accuracy (e.g. anechoic vs. quasi-anechoic vs. NFS), but on average one can expect them to converge on the same numbers, and in practice they do, as has been shown many times.

In contrast, non-linear distortion doesn't follow any widely agreed-upon standard (aside from the definition of the THD formula itself, which is a small part). You might get wildly different results depending on drive level, test signal used (frequency, shape), background noise, acoustical environment, and perhaps other factors like short-term vs. long-term distortion. Protocols are all over the place and it's rare to find two protocols that are sufficiently similar that the results can be compared directly. This is not surprising because there is still a lot we don't know about how to make non-linear distortion measurements that actually correlate well with human perception, so this area is still very much in the "experiment"/"trial and error" stage as opposed to the "standardization" stage.

I agree with everything you said, so ok, let's ditch the THD comparison and discuss how Amir is doing it. What exactly is his protocol? Is it also pseudo-anechoic as FR measurement?

I kindly ask you to put a link if all this has been described somewhere else before.

 

[617]

Everything you said is true and it does to some extent affect precision of THD measurement. Unfortunately all that in the same manner also affect frequency response measurements but I don't remember anybody complaining. In fact there have been many comparisons of various FRs made by different sources and in totally different venues.

P.S. Don't you think quoting Toole has become kind of obsolete as pretty much all of us have read the book?

The basic shape of a FR curve can be very similar between differing measurements - even moving off axis or driving the speaker louder or quieter will create very similar curves. FR curves show relative levels at different frequencies; it's rare for FR measurements made by hobbyists (or Amir) to show absolute levels, because it is expensive and a pain in the rear to calibrate everything from the amplifier to the speaker being tested to the mic to the preamp to the ADC to the software making the graph.

Unfortunately in THD we are measuring absolute levels, and the percentage of distortion changes significantly depending on how loud you play the speaker. As a result, these level calibrations matter a lot more.

Regarding Toole, I don't think we can assume we've all read his book. I haven't, I just quoted him because he summarized the situation neatly. My knowledge of distortion testing comes from my own experience, other people like @hardisj doing it, and the manuals provided by the people who make distortion testing software.

 

[QMuse]

Amir’s comment on a kind of buzzing distortion mirrors what I heard when I bought a pair of KEF Q150s on a whim. Overall that speaker struck me as having a decent frequency response but that distortion rendered them unlistenable. I guess I would describe it as a kind of harshness. It was unfamiliar to me, and I wondered if it was an attribute of the coaxial driver?

I trust his comment about distortion together with yours, but the question is how we can correlate that with distortion measurement.

 

[QMuse]

Unfortunately in THD we are measuring absolute levels, and the percentage of distortion changes significantly depending on how loud you play the speaker. As a result, these level calibrations matter a lot more.

I'd say we are measuring level of each harmonics relative to the level of main signal.

It shouldn't be confused with the fact that those relative levels usually change as the absolute level of main signal changes.

 

[SynthesisCinema]

I think the famous reputation Q100 has more to do with its great looks than sound. Measurements are not horrible but they lack the evenness that we need to see in on-axis measurements.

Overall, I can't recommend the KEF Q100. Look forward to testing their current production models.

Disappointing. Has anyone reached for you to send new Q150/Q350 or R100/R300/R3?

 

[hardisj]

Right. Distortion with Klippel's TRF module is relative to the fundamental. That's why you can show it in level (dB) or percentage. And you easily get the % based on the level differences. This happens to correlate with why you always see distortion ramp up on the low end; where a woofer's response is decreasing.

Edit: This is a good site to calculate the % distortion based on fundamental vs nth level delta.
http://www.sengpielaudio.com/calculator-thd.htm

 

[wwenze]

In before the distortion is IMD from the woofer modulating the tweeter which doesn't show up with pure tones. Wild guess.

 

[hardisj]

In before the distortion is IMD from the woofer modulating the tweeter which doesn't show up with sine waves. Wild guess.

FWIW, I tested the raw Q100 and LS50 drivers years back. Seems folks around here often ask for this kind of info. So here you go:
https://www.erinsaudiocorner.com/driveunits/kef-q100-drive-unit/
<snip>

For the 'fun' stuff, scroll to the bottom of the Q100 test and look at my testing of what happens to the tweeter response when the woofer cone is at rest, out and in.

It could be an issue depending on crossover and output level of the subjective evaluation.

 

[spacevector]

And a note on scoring: I do not have high confidence in the results.

Amir, are you saying you do not have high confidence in how the score is being calculated based on data you are publishing or are you lacking confidence in the score's formula itself?

At some point we need verification of those scores against my measurements.

Can you share some thoughts on how you would go about doing this?

 

[Xulonn]

"The coaxial driver looks great and caters to audiophile intuition of "point source being good."" Is this intuition incorrect?

We'll find out if a high-end "dual-concentric" is good when @amirm hoists a $25K, 304 pound Tannoy Westminster Royal up onto the Klippel platform for testing. /snark

Actually, it would be interesting to compare similar size, lower price-range - and much smaller - Kef and Tannoy "dual concentric" loudspeakers. England's Kef has been in business since 1961, and Tannoy (Scotland) was founded in 1926. The Westminster Royals at $50K per pair, are a horn design, so the tube amps in the picture below would likely meet the 20w minimum requirement for these monsters.

OTOH, Kef's top-of-the-line speaker is the Muon ($198K/pair, 253 lb each) in the bottom photo, which unlike the staid and classic "old Brit" look of the Tannoys, is a very modern multiple discreet driver design.

 

[QMuse]

We'll find out if a high-end "dual-concentric" is good when @amirm hoists a $25K, 304 pound Tannoy Westminster Royal up onto the Klippel platform for testing. /snark

Actually, it would be interesting to compare similar size, lower price-range - and much smaller - Kef and Tannoy "dual concentric" loudspeakers. England's Kef has been in business since 1961, and Tannoy (Scotland) was founded in 1926. The Westminster Royals at $50K per pair, are a horn design, so the tube amps in the picture below would likely meet the 20w minimum requirement for these monsters.

OTOH, Kef's top-of-the-line speaker is the Muon ($198K/pair, 253 lb each) in the bottom photo, which unlike the staid and classic "old Brit" look of the Tannoys, is a very modern multiple discreet driver design.

View attachment 53816

View attachment 53817

It took me quite a long time to notice a speaker on the upper photo.

 

[Selah Audio]

A breakup in the 1K area may be an edge (surround) resonance on the woofer. Stuffing the port would verify the cause.

 

[georgeT]

I don't think it's fair to test any bookshelf speaker with woofers smaller than 7-8" without a 60-80hz high pass filter.

Disclaimer: I own the Q350s .

 

[bobbooo]

It took me quite a long time to notice a speaker on the upper photo.

Me too - I just couldn't take my eyes off those stunning curtains!

 

[bobbooo]

And a note on scoring: I do not have high confidence in the results.

Actually I recently come across this post on the whatsbestforum by Kerem Başaran, a mechanical/acoustic engineer who says he had written some Matlab code for calculating Olive's formula. Maybe you could try to get in contact with him and see if he would be willing to share that code, or calculate the scores using it for some of the speakers already measured here, in order to provide some independent corroboration?

His member profile links to his acoustics consultancy company website (www.mtises.com), which actually has a specific page (open with Google Chrome to auto-translate) mentioning Olive's formula patent. Looks like he knows his stuff. His email address is listed as [email protected] on his website.

 

[QMuse]

Me too - I just couldn't take my eyes off those stunning curtains!

Oh, so curtains were there on the photo, too?