KEF R3 meta Measurements

[dogmamann]

Thank you for the detailed measurements.

I have a pair of "normal" R3... and dang I love these speakers.
The mid is creamy while the treble doesn't make your ears bleed like typical KEF.
As for the bass, it's plenty for me but sure not for headbangers.

My only nitpicking is,,, I hope it's a 92dB speakers lol.

What is not good about 92db speaker? That’s a very high sensitivity for a bookshelf!

 

[ROOSKIE]

That’s the frequency at which the port is tuned so you’re missing out on clean output.

No, you are not exactly only missing out on clean output.
At Fb distortion is much lower than the the same driver in a closed system as the port is now resonating and the driver is hardly moving. Here it is about 80%+port and 20% driver.
Right below Fb the system losses all control of the driver.
So, any notes just below fb with have vastly increasing HD and especially vastly increasing intermodular distortion as the cone will literally be flopping around like crazy and powerhandling will go rapidly down very soon after Fb.
It not unusual to be able to handle only a few watts before xceeding xmax just a handful of hrz below fb.

In my testing this speaker should not be run full range.
This like nearly all small driver ported speakers.

the output and distortion levels are Purifi tear so i don't see the need.

The output levels of the R3 woofer are not Puri-fi level. I saw no evidence of this driver having 10mm xmax and the like.

 

[ROOSKIE]

My only nitpicking is,,, I hope it's a 92dB speakers lol.

I am not sure if you are kidding around, but just in case not.
There are no 92db sensitive speakers with a single small 6.5" woofer and any sort of deep bass. That is not going to happen.
You can use a high sensitivity 6.5" woofer but then you must give up deep bass and excursion.
Woofers that have deep bass potential and the excursion required to play deep bass at a satisfying level are not very sensitive, especially when they have a small SD like a 6.5".

 

[ROOSKIE]

The bass could be greatly improved by making an active version.

Well yes, but there are better 6.5“ drivers than the one they use. Look at what the (more expensice) KH150 is capable of.

However probably not at 800-900€/pair (excl. VAT) the first gen actually costs (and these will slowly approach as well).
Performance is probably similar to the ~80€ standalone drivers, obviously these are custom made, but I can‘t imagine it‘s anything special in terms of performance.

You can use PEQ/DSP to increase the system extension, which is what Neumann has done. There is no magic of any kind.
You give up total output levels and dynamic potential.
You will also have higher HD/IMD at higher SPL.
The trade makes sense if used in the nearfield where less total output is required.

If you want the R3(or KH150) to play very loudly in the farfield you have to reduce the demands on the 6.5" woofers. A 6db PEQ boost at 40hrz that works in the nearfield(or at low SPL in the far) will kill the speaker if higher SPL at fairfield distances is called upon.

The other option is to use a very high excursion 6.5" driver but then you give up sensitivity (and usually have to deal with the effects of a beefy surround). In an active mode this is fine as the woofer can just use a more powerful amp. In a passive then the total system will be low sensitivity which may or may not be a design issue.

Wondering how it would perform with a Purify 6.5" driver in it. Through I would imagine KEF would never use such a "weird" looking driver since they put too much emphasis on exterior design.

It would be low sensitivity and likely need to use a passive radiator design to get a lower tuning in a small box. The weird surround of the Purifi is not really necessary in a 3-way. It would just need a well designed but very beefy surround.
The Purifi is designed for max-high performance, small 2-way application.

I know KEF does their drivers in house, but would have loved to see a Purifi 6.5 or "even" an SBA Satori or Scsn-Speak Revelator driver. Given what the BMR monitor can do in a similarly sized cabinet

I have the BMR, it has a nice bass extension. For higher SPL the driver can not handle it.
For good output at SPL it should be used with a 70 or 80hrz high pass and subwoofer augmentation.
There is no benefit to crossing lower as the cabinet is large so it does not control the woofer well from 40-70hrz at higher output levels.

While I think many folks will be satisfied with the BMR's output levels, if you listen to bass heavy music at medium high to high levels in a decent sied space, the speaker will be very stressed.

Again remember that 2, 6.5" woofers only have the same output level of a comparable x-max 8.5" woofer. This can be considered essentially always the case. The SB acoustics woofers used in the BMR have between 3-6mm Xmax depending on the model/approach for generating specification. That means they output maximum SPL levels of a single 8.5" woofer with 3-6mm x-max. (and they will require more power to do it as the 8.5" will be usually be more sensitive all things being equal)

There is no magic of any kind.

Even the Purifi 6.5" has up to 10mm of x-max depending on the modeling approach for specing xmax. That means 2, 6.5" $440 purifi woofers ($880 for two) have the output levels of a typical, single 8.5" woofer with 8-10mm of xmax. That puts the Purifi in perspective eh'? Those are low sensitivity as well, though for the design type they are good.
This should put in perspective how much better any typical decent subwoofer is vs the extremely expensive Purifi. The Purif is a cool concept though and well engineered for what it is. It is just that what it is very niche and very expensive without enough real world gains in most cases.

 

[Alexx]

@Snix: I totally agree the old R3 has a fantastic upper midrange and good bass.

I wonder if "by ear" how much difference can be perceived between R3 and R3 Meta since instrumentally it does not seem to be so gigantic.

Another humble consideration of mine, is it possible that the new R3 Meta series seems to be a little more "open" on the mid-highs because they have attenuated the low range and consequently it covers the high frequencies less?

I understand that the Meta series has lower and less pronounced bass than the previous series correct?

Thank you very much and congratulations for the review.

 

[ROOSKIE]

No doubt that the KEFs are an excellent speaker, the Helios is a superlative speaker. There is a tendency, and I'm not saying you are one of those, to be dismissive of kit speakers. There is a many page thread that pit the Philharmonic BMR's (available as a kit for 1000USD) vs the R3's and while there is not a clear winner, the BMR clearly hold their own and stomped the R3's at a recent unblinded battle of the stand mounts. In fact they pretty much stomped all of the competition. Now those do have a retail outlet and so are not exactly typical of the kits, but guys like Jeff Bagby and Dennis Murphy to name a couple are scarcely amateurs. Theres a great deal of simulation software that was written by avid amateurs--some of which is now commercial. Now so far as I know none own a Klippel, but many exceptional were designed and are being designed without one.

Have you heard the Helios? Hard to call it superlative without some subjective experience and honestly some real Klipple testing.
I have dabbled in interest in that speaker myself but that is a big commitment for something you will never be able to 'return'.
I am a DIY hobbyist. I have definitely built stuff I didn't truly love, including some expensive kits.

I have mad respect for Jeff Bagby(RIP) however the one speaker I made of his was not exactly what I would want now and did not measure all that well. (The PartsExpress Tango) I was able to drop the passive crossover and turn them into actives that measured and sounded much better. Now, I am using the drivers, enclosure and parts in various other projects so the $$investment still has legs and fun times.
That said, the risk of not liking the speakers is real.

So sure KEF, B&W, Harmon Group have the tools at their disposal to make high value speakers, but remember all but the cost no object are the product of many compromises that the DYI crowd are not subject to--say choosing to use air wound vs ferrite core inductors, spending $50 dollars more for a woofer, adding overkill bracing, using automotive finishes, and so forth. Having active speakers makes the development so much faster than the days of yore (pre-SPICE must have been a bitch) as xo's can be changed on the fly and otherwise mismatched drivers made to work seamlessly. It's a very cool era for DYI--great kits, unbelievably good drivers, powerful, often free software and the internet makes for a vastly different space than the old days.

How much should the screws cost? I say this light heartedly but really it has truth to me.
Parts are parts. As I am sure many realize, a person can make superb speakers with lower cost parts and bad speakers with expensive parts.

Keep in mind as well that many DIY speakers and parts are now retail/commercial products. With all the marketing and 'design' trappings in play. Some of the best design choices for DIY are still the most pedestrian.(look at the BMR, nice basic SB acoustics woofers and a very inexpensive midrange from the OEM market. Pricey tweeter though.)
Overkill bracing and more expensive crossover parts is not exactly representative of excellence or a guarantee of performance. Nor is changing active crossovers 'on the fly', each change really has to be considered & tested/explored - it is clearly faster than passive parts but it is still a process not to be underestimated.

Manufacturers in 2023 can engineer extremely good quality parts for crazy low prices and often some of these parts will appear cheesy in the hand, but are in fact quite well engineered(and vice versa with very beefy parts sometimes just being average performers despite $$). You simply can't judge a book by its cover.
Many large scale manufacturers can make extremely high performance parts at very low costs, even if ordering them outsourced from an OEM. Because it is 2023 and computer modding and other engineering tools are so powerful a very pedestrian looking driver may in fact be superbly engineered yet cost very little $$ to make. Even high value items like stamped steel baskets and even plastic baskets can outperform cast baskets of old.

In any case what a DIY buyer pays for the same performance in a driver that a large manufacturer pays is sometimes very far apart - as in I might pay $75-200 for a woofer of the same performance quality that costs JBL or KEF $10-20.

All that said I do advocate for DIY as the best way to learn and for the right person is it fun - truly something worthwhile to do. At the higher end of DIY you can save money as well.(I don't think you can save much money anymore with entry and mid level DIY project vs 'top of price class retail-when on sale prices')

If you make those Helios and feel crazy - send them in for a Klipple test. I am sure many would be curious. I know I am, they sure look like a possibly great speaker.

 

[abdo123]

The output levels of the R3 woofer are not Puri-fi level. I saw no evidence of this driver having 10mm xmax and the like.

Distortion wise they're close to each other at 96dB.

 

[dogmamann]

I understand that the Meta series has lower and less pronounced bass than the previous series correct?

it’s not mentioned that way, he didn’t mention it compared to the previous version but generally compared to all bookshelf speakers. You are not loosing anything with the meta but gaining quite a bit in terms of soundstage. If you have personally listened to the previous one, it really had a narrow soundstage for their size. This definitely is a win for me. If you are looking for reasons, do justify the purchase of the old version, soundstage should not be your priority.

Between ls50 and ls50 Meta, I felt the sound less boxy somehow. R3 already was good but still had a tint of that character, but if that’s gone, r3 meta would become one of the top dogs in the category. Best is end of the day subjective and preference matters here!

 

[JRS]

No not dismissing kit gear at all. I have “kit” amps, but they are measured to the last detail. Just how I like it, others might have other preferences. As always there is good (eg Hypex, Purify) and bad kit gear (eg the GR speakers). I guess if one is interested, one can always build the kit speakers and send to to Armir. But we are getting a bit OT.

Indeed we are getting OT--apologies to the OP. One last comment: some of the kits are very well documented--Troels Gravesen is a good example as is John Krutke, Jon March, Jeff Bagby, Dennis Murphy and a number of others. Even without spins a fair amt of info is present with re to FR, THD, off axis behavior, dynamic compression, etc.

 

[Alexx]

Thanks a lot for the info.

I misunderstood the bass speech, sorry.

I absolutely don't have to convince myself that the R3 series is as good as the new "Meta" surely some improvements have been made.

As far as the soundstage is concerned, it doesn't matter much to me, I believe that even the old model can have some good characteristics even if with some small shortcomings......

Certainly later not excluded that I will update with the R3 Meta.

For the moment, it doesn't seem to me that there is such a shocking difference that it justifies a difference of around 700 euros.

Other direct comparisons can be useful for all users who have some doubts.

Thank you very much for the information.
See you soon.

 

[sweetchaos]

R3 Meta, based on these preliminary measurements, solidifies the concept that this is a world-class type speaker.

Now, we can confidently have someone send in a sample for review to either Amir or Erin, to see the precise Klippel NFS measurements. *wink*

 

[cavedriver]

KEF's precise directional control is truly amazing.

Its predecessor was good enough, but it's even better than I expected.

I'm afraid I don't understand this statement. The KEF's appear to have a fairly narrow beamwidth, getting as low as +/-30 degrees starting at just above 4k for -3 dB and +/-40 degrees for -6 dB. Isn't 50~60 degrees considered "good", while these numbers could be considered perhaps "average" but certainly not "amazing"?

 

[dogmamann]

I'm afraid I don't understand this statement. The KEF's appear to have a fairly narrow beamwidth, getting as low as +/-30 degrees starting at just above 4k for -3 dB and +/-40 degrees for -6 dB. Isn't 50~60 degrees considered "good", while these numbers could be considered perhaps "average" but certainly not "amazing"?

He meant it has an amazing narrow soundstage. Kef is one of the speaker which is meant for people who wants narrow soundstage. It’s a criteria for some. You cannot say only 50-60 is good. Everything is good, it depends on preference.

 

[cavedriver]

He meant it has an amazing narrow soundstage. Kef is one of the speaker which is meant for people who wants narrow soundstage. It’s a criteria for some. You cannot say only 50-60 is good. Everything is good, it depends on preference.

Except that virtually every other KEF reviewed on this site runs around +/-50 degrees at -6 dB. My point is this speaker is unusually narrow, and while I appreciate that directivity to some degree is apparently a matter of taste, there must be a thing as too narrow. Effects like "small soundstage, "narrow sweetspot", "very sensitive to positioning" etc. will be said. Not one of the top 20 or so reviewed speakers on this site have dispersions this narrow. Perhaps the scoring system favors wider dispersion?

 

[abdo123]

Except that virtually every other KEF reviewed on this site runs around +/-50 degrees at -6 dB. My point is this speaker is unusually narrow, and while I appreciate that directivity to some degree is apparently a matter of taste, there must be a thing as too narrow. Effects like "small soundstage, "narrow sweetspot", "very sensitive to positioning" etc. will be said. Not one of the top 20 or so reviewed speakers on this site have dispersions this narrow. Perhaps the scoring system favors wider dispersion?

Actually it favours narrow dispersion, it has been discussed thoroughly on the forum.

 

[thewas]

Actually it favours narrow dispersion, it has been discussed thoroughly on the forum.

It favours linearly increasing increasing directivity while it penalizes constant directivity.

 

[Alexx]

I have no idea how many degrees of openness a live event must have...

I can imagine that if an instrument is on the left and the singer on the right, I think he should remain in the same position even when listening directly from the speakers (therefore voice and instrument shouldn't invade the whole scene otherwise they denaturalize the rendering).

I don't know if I explained myself well, it's not easy with the translator.

I love the Kef sound and like any other brand I believe a little compromise should always be there... hard to find the perfect speaker.

I also liked Sonus Faber Sonetto 2 a lot, but I like 3-way speakers too much and Kef is one of the few I can afford in this price range.

In about 1-2 months the new Meta series will also be available in Italy, I will certainly make a direct comparison by taking mine to the shop to listen together, if it is worth it I will change, this cannot be excluded, otherwise I will keep mine for at least one year.
Thanks a lot See you soon.

 

[3125b]

Now, we can confidently have someone send in a sample for review to either Amir or Erin, to see the precise Klippel NFS measurements. *wink*

Erin said on Youtube that he would resume testing in the near future.
I'd be really interested to see his linearity and IMD tests on these and how close they might be to the Reference 1 Meta.

 

[cavedriver]

Actually it favours narrow dispersion, it has been discussed thoroughly on the forum.

Which makes it doubly fascinating that the top scoring speakers all have wider dispersion than this one. I searched and I couldn't even find one "outlier"- a speaker with less than 45 degrees of smooth dispersion (at -6 dB). If narrower dispersion is favored by the scoring this speaker should have some kind of crazy high preference score, especially for the "with a sub" score.

 

[BrokenEnglishGuy]

Which makes it doubly fascinating that the top scoring speakers all have wider dispersion than this one. I searched and I couldn't even find one "outlier"- a speaker with less than 45 degrees of smooth dispersion (at -6 dB). If narrower dispersion is favored by the scoring this speaker should have some kind of crazy high preference score, especially for the "with a sub" score.

Wide dispersion in small rooms are a mess. Thats why there is no point for that.
Wide or narrow dispersion? You need to know the room first if its big enough for a very wide dispersion