KEF R3 Speaker Review

[CtheArgie]

@amirm, is it possible that you have such a strong preference for the 100Hz-200Hz boost speakers compared to those without it because of the music you use to listen to them? If I recall (sorry if I am wrong), you have a lot of bass heavy music in your choices. What would be your reaction to them if you listened to string quartets or just acoustic guitars and voice? Can these speakers then be well complemented with a pair of subwoofers given their good behavior on the rest of the frequencies?

 

[infinitesymphony]

While we're on the subject of Amir's Ears (new podcast title?), I would be curious to know Amir's impressions of how his listening room generally interacts with speakers. Does it tend to emphasize or cut any frequencies? Any long decay?

Maybe we could take an impulse response of Amir's listening room and apply that to the measurement data to see what a speaker might look like there, then correlate it with his listening impressions.

 

[digitalfrost]

So if we compare this with the KEF LS50, it is a somewhat familiar picture. Slightly recessed midrange, with a overemphasized highs. Now, I own LS50 and I love them, but I think they greatly benefit from DRC, even above Schröder frequency. The highs might the reason why.

That said. KEF coax + DRC = some of the best results I heard so far. I need to listen to some Genelecs some of these days.

 

[thewas]

KEF R3 is -3 dB @ 65 Hz and has a bump in the treble on a mostly flat FR graph.
Revel M16 is -3 dB @ 60 Hz and has a bump in the bass and slightly rolled-off treble on a mostly flat FR graph.

Maybe just a preference for warmth / bass emphasis which Amir mentions as a possibility in his impressions.

Here are the predicted in room responses of both, although at the (too) high listening distance of 12 feet for such speakers the sound power part should dominate even more, I matched them level wise at the ground tones region 200-500 Hz:

We see 2 main differences, one is the bass boost of the M16 which is often used for small speakers to compensate their early bass drop and make them sound also "larger" then they are and possibly also done for the US market where rooms are usually bigger and often have not solid walls like in Europe were the R3 was voiced. The other is the mid region were a higher level of the Revel makes something sound more "lively".

That said, bass behaviour should imho be equalized to a room or optimised by placement as otherwise just a random combination of loudspeaker, room acoustics and geometric placement will "win" although it shouldn't be objectively generalised as superior and we know from Toole that bass plays the biggest role in the subjective ranking of a loudspeaker. Also we shouldn't forget that when just one loudspeaker is used vs. a stereo pair, then its tonality is perceived as more bass shy as at the lower frequencies the addition of 2 loudspeakes is closer to 6dB (coherent signals) vs. to higher frequencies where its closer to 3dB (non-coherent) so a single loudspeaker with a bass boost will sound more neutral than a neutral one.

 

[JohnYang1997]

Looking good. Maybe I'll have a listen one day.

 

[tecnogadget]

On axis, R3 (red) vs M16 (blue). As you can see R3 is much more linear.

View attachment 54022

This is my grain of salt and, by the way, thanks QMuse for your post.
Usually, people are not used to listening to linear or natural frequency responses and may think it "sounds boring" or "lacks life". The same applies to a professionally calibrated screen next to one with more saturated colors...90% of customers would choose the latter in a showroom.

The thing is that you liked that colored punchy sound in the demo room, or that flashy screen...you set up both on your home and think they are great...but gradually after time you find yourself tweaking because find out that with some specific movies or music find some flaws...

Most of you must be aware of how not level matched A/B benchmarks usually bias for punchier bass, even 2 speakers with a similar response if one is a little bit hotter in levels it feels MUCH better than the other one (same applies to A/B testing DAC's)

QMuse plot of both R3 and M16 demonstrates it. Both are good performers with overall smooth FR but just different tendencies/tastes. M16 has that bump in the lows and a slope down in the highs, so it much resembles our desirable "in-room" target response. R3 are more neutral linear with extended highs, so yeah they could be bright to some folks, but who cares?
Because I would cross both speakers to a subwoofer and then apply parametric EQ (placing sub in the highest output spot of the listening room so you can lower any peaks and ringing). The ideal in-room would be like the Genelec 8341A but with some gain in the 20hz to 100hz region (leave it to taste between 3 to 5 db) like a "hard knee" curve.
This is super easy, you either make little Eq to lower in the 3k to 10k region for R3, or increase a little bit the 200hz to 600hz in the M16 and tame the 5k rise. What matters is having wide dispersion and cohesion so this tweak does not affect the end result.

To summarise, going back to where I talk about tweaking because you don't like the system. If your in-room ends up as it should, after listening for a long time you will find any genre or recording sounds good and start discerning bad recordings instead of bad speakers .
That's something it can't correlate to informal short listening sesions.

 

[thewas]

While we're on the subject of Amir's Ears (new podcast title?), I would be curious to know Amir's impressions of how his listening room generally interacts with speakers. Does it tend to emphasize or cut any frequencies? Any long decay?

Maybe we could take an impulse response of Amir's listening room and apply that to the measurement data to see what a speaker might look like there, then correlate it with his listening impressions.

I would prefer some MMM measurements around his listening position.

 

[richard12511]

These last few KEF reviews have showed speakers that look technically superior to the JBL 705p, and this one is even better than any of the Revels measured so far, but it seems you subjectively much prefer the JBL and Revels. You gave them glowing recommendations despite their seemingly worse objective measurements. Isn't it kinda the point that we should be able to predict listener preference based on measurements?

You not recommending the Q100 really surprised me. To my eyes it looks as good or better than the 705p, and it does it at a fifth of the cost. Seems like an easy recommendation.

Can't wait til we can start doing blind listening tests to compare to these measurements. If you keep this up, pretty soon you'll have more than the 70 speakers sample size that Harman used. Would love to see y'all come up with your own "Preference" formula based on your own measurements and listening tests, and I feel like it's doable.

When you said you were gonna start measuring speakers, I had no idea you'd be getting so many measured so quickly. This is awesome!

 

[laurelkurt]

I linked to your "glowing objective measurements" review in Kef Enthusiasts Facebook Group. I hope they don't all come looking for your head. I can remove it if you like. Got a like from Jack Oclee-Brown, the designer. I certainly hope whomever's R3 warranty hasn't been violated by sending these in!

 

[PBH]

Thank you for the great review. I must say I am not really a fan of the pinkish-gold drivers with the wood veneer.

It seems to be a polarizing colorway. Personally, I think it's one of the neatest speakers that I've seen years.

 

[ROOSKIE]

It seems to be a polarizing colorway. Personally, I think it's one of the neatest speakers that I've seen years.

In person I personally like it, while oline the images don't seem to capture it the same way. I prefer the black though. Beautiful speaker.

 

[ROOSKIE]

I have to agree on there being something about the Revel line that the KEFs lack. I own KEFs, and have heard a lot of them and thought they were 'good', but a while back I was in a local store and happened to hear some M106s playing. They're something else. I would love to have that 'something else' quantified.

I am curious how other speakers using SB acoustics drivers fare here.
Especially this satori kit.
https://www.madisoundspeakerstore.com/speaker-kits/satori-ara-2-way-speaker-kit-pair/

 

[amirm]

I would love to chip in a bit. I have a small ask though - can you please apply the broad EQ you mentioned that would help the speaker and redo the listening test against the Revel M106?

I can't. The measurement system is under the control of Klippel software and I have no way of messing with datapath. Others can easily simulate that though.

 

[infinitesymphony]

Also we shouldn't forget that when just one loudspeaker is used vs. a stereo pair, then its tonality is perceived as more bass shy as at the lower frequencies the addition of 2 loudspeakes is closer to 6dB (coherent signals) vs. to higher frequencies where its closer to 3dB (non-coherent) so a single loudspeaker with a bass boost will sound more neutral than a neutral one.

This sounds like pan law, but wouldn't that only apply to individual channels of a stereo mix that have not been summed to mono?

e.g. If you fed a mono mix through two KEF R3s and two Revel M16s, do you think the impression would change?

 

[thewas]

This sounds like pan law, but wouldn't that only apply to individual channels of a stereo mix that have not been summed to mono?

Usually the bass in most recordings is mainly mono to a high part for several reasons, some of them even historic (vinyl).

e.g. If you fed a mono mix through two KEF R3s and two Revel M16s, do you think the impression would change?

Imho woudn't even need to be a mono mix for above reason.

 

[infinitesymphony]

Usually the bass in most recordings is mainly mono for several reasons, some of them even historic.

Definitely, that's what I thought you meant. Shouldn't this issue go away if Amir sums to mono before outputting to the speaker? I suppose I assumed that's what was happening.

 

[SplitTime]

It gives -some- insight in unit-to-unit variation. Apart from that.. not really.

Two samples just isn't enough to draw any conclusions about unit-to-unit variation. Depending on the sigma of variation you're looking for at least 100s of units would need to be measured.

@amirm - please continue measuring new units instead of repeats.

 

[thewas]

Definitely, that's what I thought you meant. Shouldn't this issue go away if Amir sums to mono before outputting to the speaker? I suppose I assumed that's what was happening.

If I understood correctly he compares always one loudspeaker vs. the other but when you listen to a single loudspeaker it will appear a bit bass shy compared to the tonality you are used to from your usual stereo pair listening, so a single loudspeaker with a bass boost will sound more normal and less bass shy compared to a neutral one.

 

[BYRTT]

.....Discussion and Conclusions
The objective measurements will nail the Olive score no doubt. And they present for the third time a conflict with my subjective listening impressions. Of course my subjective evaluation is much less reliable. I like us to allow some allowance for them to sink in though. Olive's latest tests shows people like to hear more bass than originally though. So could this be behind my preference for speakers like Revel M16? ...

I am really starting to think the 100 to 200 Hz region plays a much stronger role than we think in subjective sound a speaker produces. The other factor is not letting the higher frequencies dominate the mid-range. As I noted in the review, broad deviations in the measurements, despite their low level, may have a much larger subjective difference.

At some point we will have to reconcile these differences, either setting me straight on my subjective evaluations being wrong, or us not knowing all that Harman knows about good speaker sound. Let's remember that they won't release a new speaker unless it passes double blind listening tests against its competitor. No other score allows them to skip this test. Components are tweaked until they achieve this. So one wonders if this is not released to public...

Great thanks nice review and spin data, we also interested eat what you sense for subjective part at least i do and in you try to find explanations why M16 get golf panther and R3 plus 8341A dont get their expected football panther i made below toggle of 3 second animation camparison for M16/R3/8341A, maybe its too much show there but think its a good stare in Spinorama have included PIR in the upper orange curve and also a rough curve of Toole's trained listener preference from first edition book where M16 is coming most close to that curve. Thanks if you will take a stare and see if any directivity or Spinorama data makes sense to listening test, in the lower animation they filtered 8th order to telephone band like bandwidth and also smothed flat as pancake on axis which think combined the filtering is a good effect ala normalize thing.

Click inside visual to get clear resolution to directivity patterns:

EQed flat on axis and filtered BW 8th order stopbands @80Hz / @7kHz, click inside visual to get clear resolution to directivity patterns

 

[amirm]

@amirm, is it possible that you have such a strong preference for the 100Hz-200Hz boost speakers compared to those without it because of the music you use to listen to them? If I recall (sorry if I am wrong), you have a lot of bass heavy music in your choices.

Answering the second part, the music is not bass heavy. I use those clips for headphone amp testing since they run out of power and distorted. The list I have here is a mix of "classic audio show clips and my own favorites that sound wonderful on my system. Here are a few of them: