KEF R3 Speaker Review

[fmplayer]

HI!
Don't trust What Hifi reviews, it's all subjectivist stuff, no measurements. Check out more serious sites as this one or
Stereophile. Subjective test mean nothing without scientific measurements to compare with.

Aswell as objectively good means nothing if subjectively bad. Kef R3 or UD501 are good examples of objective and subjective quality. Alas, there's a lot of things to say about the discrepancy between perception and measurements, so my method is to make a short list based on objective measurements than pick smth from the list based on listening sessions.

HI!
Forget about Atoll amps, they're badly built and probably dangerous! Check out this French report from a repairer, who is not tongued-tie:
https://hifirep.jimdofree.com/amplificateur-atoll-in100/

In France, Atoll amps are well promoted, probably because... They are made in France lol

... As well as US stuff is promoted in the US or UK stuff in UK . At least, it's not chinese . BTW, the report is about the old model. The new one seems better although I still find the radiators rather tiny for 100W. For the record, I live in France and didn't hear about any safety issue with Atoll stuff

I had same amp setup than you(Teac 501 series, all of them) but with Paradigm Prestige 85F.
I went to Audiophonics Hypex nc400 based one(1500 euros) a year ago , and IMHO, got a small improvement, with tighter bass and less distortion at very high volume(lot of power reserve, check out Amirm's test) . Now, you can get an Audiophonics purifi one for exactly same price, so get it! . HA-501 is a good headphones amp, with lot of Umph and power, but not as transparent as recent and cheaper ones, probably same thing as a preamp(Amirm tested IT, ).

Do you use the HA501 as a preamp ? it makes only 300mV, which is on the low side to drive decently a Hypex or Purifi module, even with the 25dB gain.
For the record, I also have the PD player, all high WAF stuff.

And yes, I'm studying Audiophonic's offer (it's a french company it seems ).

 

[batfunk]

I agree with your method, scientific measurements allows to make a short list:
Topping dx7 pro performs brilliantly but I reselled it.
That's why you must avoid Atoll products, no measurements anywhere. Hmmm, It's very doubtful Atoll renewed entirely their schematics used for decades. Repairer seems to be a skilled technician, I'm not, I trust him.He blames often Nad also for cheap components Retailers don't give a damn about implementation , they sell, period. Made in France, as made in USA or Canada, is a merchandising argument,useful to sell averages products overpriced, and it works for good products (Focal) and bad ones.
It's your money after all... ;-)
I never used Ha-501 as preamp, no need with my previous AI-501.I only still use Pd-501,great cd player!
My SMSL m200 works fine with my Audiophonics stereo Nc400 Amp, you've got power Rca output of 2.1 vRms and 4.1 vrms with Xlr output.
And if product performs well, I don't care if it's chinese, Indian or french stuff.
Today, most Competitive dacs are made by Chinese brands(SMSL, Topping).
My speakers are Canadian., my power amp was assembled in France with Chinese parts engineered by a Dutch company(Hypex), my headphone and amp are Chinese (SMSL),my headphones are made in Usa(Grado) and my cd player was made in China for a Japanese company (Teac) who cares?

 

[fmplayer]

Topping dx7 pro performs brilliantly but I reselled it.

Good to know

It's your money after all... ;-)

Atoll is no longer part of the equation

I never used Ha-501 as preamp, no need with my previous AI-501.I only still use Pd-501,great cd player!

Yes, and you can rip a few CD's on a single DVD as WAV files, without error correction and so on. Less space and subjectively much better.

My SMSL m200 works fine with my Audiophonics stereo Nc400 Amp, you've got power Rca output of 2.1 vRms and 4.1 vrms with Xlr output.

Good to know

And if product performs well, I don't care if it's chinese, Indian or french stuff.
Today, most Competitive dacs are made by Chinese brands(SMSL, Topping).
My speakers are Canadian., my power amp was assembled in France with Chinese parts engineered by a Dutch company(Hypex), my headphone and amp are Chinese (SMSL),my headphones are made in Usa(Grado) and my cd player was made in China for a Japanese company (Teac) who cares?

Well, I do care. Chinese people won't pay for my retirement income with their work. French will. There's a lot of other reasons, but this thread is not the place to debate. But what we can discuss is a general purpose quote: "Everything comes at a price"... Even amplifier distorsion figures ... albeit in a specific thread

 

[batfunk]

If you intend to go to Audiophonics stuff, you have to know they provide also an excellent aftersale service: They changed my Nc400 amp faulty power switch And sent it back to me the same day they received it!
Advance acoustics seems to be also a serious french company(as Focal) but I found no measurements, so....

 

[fmplayer]

If you intend to go to Audiophonics stuff, you have to know they provide also an excellent aftersale service: They changed my Nc400 amp faulty power switch And sent it back to me the same day they received it!

Good to know. After reading Archimago's recent test of NC252MP, my intention is to buy Audiophonics' version for 470EUR, which is a bargain. Don't even think about DIY'ing a class D amp by yourself. I don't listen very loud, so 250W/4ohm should be OK. I'm just waiting for my R3's to come before ordering. Just wondering if I really need that extra gain (15dB). Nelson Pass released a one stage low distortion no NFB balanced Mosfet circuit that might be suitable. Time will tell and report to come

Advance acoustics seems to be also a serious french company(as Focal) but I found no measurements, so....

Saw it but never heard it, so ...

 

[batfunk]

Good to know. After reading Archimago's recent test of NC252MP, my intention is to buy Audiophonics' version for 470EUR, which is a bargain. Don't even think about DIY'ing a class D amp by yourself. I don't listen very loud, so 250W/4ohm should be OK. I'm just waiting for my R3's to come before ordering. Just wondering if I really need that extra gain (15dB). Nelson Pass released a one stage low distortion no NFB balanced Mosfet circuit that might be suitable. Time will tell and report to come

Saw it but never heard it, so ...

If I remember, you can modify gain with a switch inside my nc400 amp. Maybe there is the same in the nc252 version. Ask question to Audiophonics. If so, ask to them if you need to change gain on amp with your new Kef R3 acoording your listening level.

 

[fmplayer]

If I remember, you can modify gain with a switch inside my nc400 amp. Maybe there is the same in the nc252 version. Ask question to Audiophonics. If so, ask to them if you need to change gain on amp with your new Kef R3 acoording your listening level.

You're right, but If my understanding is correct, you can just lower the gain with by passing the input buffer. Nominal gain is around 26 dB, and you can lower to 13. Same for Purifi

 

[BrokenEnglishGuy]

The speaker seems to measure fine, but fails the listening test. Could intermodulation distortion be the culprit here? It is the Achilles heel of coaxial drivers.

for what i can see it's only the FR, but you can EQ for add bass in 100-200hz if you like that, but if you use subs you will want the default bass of R3's
IMHO

 

[fmplayer]

Thanks to COVID and Brexit, I'll have to wait 'til end of February to get my R3s home. Pfff.

BTW, when I read that the R3s failed the listening test (@nerdoldnerdith), it shows all the subjectivity of listening ...
Before I purchase smth, I always try 3 times, to make sure I'm not fooled. If I'm not convinced at whatever stage, I stop the trial immediately, considering I don't have to adapt to something I don't really like. "Like it or leave it", "Convinced even at third sight", call it like you want. The R3s passed the test, even once with a so-so amp, contrary to the Revels. Go figure ...
... Or maybe they don't "beautify" things enough ... In the real world, raw (unprocessed) music isn't always pleasing, it depends of the quality of interpretation. FWIW

 

[nerdoldnerdith]

Thanks to COVID and Brexit, I'll have to wait 'til end of February to get my R3s home. Pfff.

BTW, when I read that the R3s failed the listening test (@nerdoldnerdith), it shows all the subjectivity of listening ...
Before I purchase smth, I always try 3 times, to make sure I'm not fooled. If I'm not convinced at whatever stage, I stop the trial immediately, considering I don't have to adapt to something I don't really like. "Like it or leave it", "Convinced even at third sight", call it like you want. The R3s passed the test, even once with a so-so amp, contrary to the Revels. Go figure ...
... Or maybe they don't "beautify" things enough ... In the real world, raw (unprocessed) music isn't always pleasing, it depends of the quality of interpretation. FWIW

I speculated that they failed the subjective listening test because of intermodulation distortion, which is a problem known to plague coaxial drivers. It wasn't measured in the review.

I tend to believe that whatever we can hear in a speaker can be measured as well. Subjectivists will say that measurements don't explain everything. This is true if you have a limited set of measurements. However, the solution is more measurements to get a better understanding, not chalking up the disparity between what we hear and a limited set of measurements to magic or hidden variables.

 

[q3cpma]

I speculated that they failed the subjective listening test because of intermodulation distortion, which is a problem known to plague coaxial drivers.

Not really in 3-way coaxials, see Genelec's Ones white paper for some measurements.

 

[BrokenEnglishGuy]

I think is only because of the FR, if the speakers have 7 FR for user taste, sure the r3 will like more in subjective listening.. the people who know how to do a manual EQ care more about the measurements and bla bla, and less interest subjective listening
imho

 

[tuga]

I speculated that they failed the subjective listening test because of intermodulation distortion, which is a problem known to plague coaxial drivers. It wasn't measured in the review.

I tend to believe that whatever we can hear in a speaker can be measured as well. Subjectivists will say that measurements don't explain everything. This is true if you have a limited set of measurements. However, the solution is more measurements to get a better understanding, not chalking up the disparity between what we hear and a limited set of measurements to magic or hidden variables.

The Kef’s directivity is narrower than what I believe Amir is used to and enjoys.

 

[Chromatischism]

I think is only because of the FR, if the speakers have 7 FR for user taste, sure the r3 will like more in subjective listening.. the people who know how to do a manual EQ care more about the measurements and bla bla, and less interest subjective listening
imho

EQ can substantially improve subjective listening so I disagree.

 

[BrokenEnglishGuy]

EQ can substantially improve subjective listening so I disagree.

That is what i'm saying.. If you have differents FRs ( EQ ) there is a large gap between the audience you can get
If you know how to EQ, well you can set the FR at your taste.

If you have 3 present:
high bass
neutral
bright highs
The people gonna choose different preset for his taste, but the speakers only have 1.

 

[thewas]

The Kef’s directivity is narrower than what I believe Amir is used to and enjoys.

And as the 1985 Toole experiments have shown, narrow directivity gets a much lower preference in a mono (single loudspeaker) than in stereo listening test.

 

[fmplayer]

I speculated that they failed the subjective listening test because of intermodulation distortion, which is a problem known to plague coaxial drivers. It wasn't measured in the review.

I tend to believe that whatever we can hear in a speaker can be measured as well. Subjectivists will say that measurements don't explain everything. This is true if you have a limited set of measurements. However, the solution is more measurements to get a better understanding, not chalking up the disparity between what we hear and a limited set of measurements to magic or hidden variables.

I'm an engineer, so go objectivism !!! Measurements explain what is related to what they measure, no more no less. That being said, we are in search for measurements that explain subjective feelings, because our senses (hearing for this matter) give us a clue which does not correlate with what we actually measure. For example, we had to wait for Otala's TIM work to explain some bad behaviour of yesteryears amps. IMO, we don't have yet the equivalent in loudspeaker measurements, since most of them are static (FR, THD, directivity, you name it). I strongly believe that a measurement protocol able to show dynamic driver (thus loudspeaker) performance will explain things like boring loudspeakers. I often use the image of 2 musicians playing the same music, on the same instrument, but with different results according to their individual technique (let's forget about their feeling differences, you get my point).

So, is a loudspeaker boring because it has no particular flaws (what is sometimes said about class D amps like nCore or Purifi), including the unmeasured for the time being dynamic domain, or is it more entertaining due to a pleasing distorsion of some kind (like tube amps with their H2, or Bob Katz's H2 generator).

OTOH, ear is a fabulous detection instrument, which has 2 flaws:
- its inability to quantify things and show the results (you can't draw a frequency curve from what you hear, can you)
- interpersonal variability (some musicians hear the difference between 3 pitches of the A note played by tuning forks, but most of us are incapable of that)

So, due to the temporary lack of full measurement set, one has no other choice but to mix quantitative measurements with hearing evaluation, for better or for worse. At least, one can be certain to pick something that pleases him, no matter if it is objectively justified or not.

So, let's find some new measurements to further correlate with our feelings. Maybe the subject of a dedicated thread ?

 

[Chromatischism]

Maybe the subject of a dedicated thread ?

Definitely beyond the scope of this thread and maybe even this website. People are moved by different things so I don't even know how you would nail down the metrics.

 

[fmplayer]

Definitely beyond the scope of this thread and maybe even this website. People are moved by different things so I don't even know how you would nail down the metrics.

Beyond the scope of this website ? Mmmmhh AudioScience ... Lots of people with technical background and interested in progress of audio scientific aspects, oportunities for ideas and knowledge sharing, ... I would argue.
What does the Boss (@amirm) think ?

 

[infinitesymphony]

So, let's find some new measurements to further correlate with our feelings. Maybe the subject of a dedicated thread ?

What happens when two people have opposite feelings about the same equipment?

The best we can do is define targets (preference curves, in-room response slopes, etc.), that sound good to most people and aim for those as a starting point. It will be up to you to determine what you like, and the measurements will be there to help you figure out why.