• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

KEF R3 Speaker Review

QMuse

Major Contributor
Joined
Feb 20, 2020
Messages
3,124
Likes
2,785
Remember, revenue is not the same thing as profit. M cars are not built from scratch so they benefit from development of the non M versions.

Oh, I'm well aware of the difference. With BMW highest profit margin is with their cheapest models, not the M series, and that is usually the case.
 

tecnogadget

Addicted to Fun and Learning
Joined
May 21, 2018
Messages
558
Likes
1,011
Location
Madrid, Spain
The Reference series in incrementally better in virtually every way. It seems to me though that in terms of the most important metrics, it's 90%+ there already with the R series, at a quarter of the price (which TBH is more a compliment to the R series than a criticism of the Reference series).

Totally agree. The Reference is simply put...THE REFERENCE ;).
I love the fact that R series are just so close to that performance at an affordable price point (in comparison with the many times' expensive sibling).

"The Reference" measurements are textbook exemplary (well let's wait until Amir fact checks that one day). But I don't doubt for a second that you can A/B an R3 with perfect calibrated room setup + Dirac + 2 Subs and the vast majority of listeners won't really appreciate night and day differences.
 

stren

Active Member
Forum Donor
Joined
Jan 9, 2020
Messages
253
Likes
298
I still hope that they will do an active R3. Or instead maybe a mini bookshelf sized blade that is active - keep the same style and mid range driver, but have two side mounted 6.5" bass drivers.
 

jhaider

Major Contributor
Forum Donor
Joined
Jun 5, 2016
Messages
2,822
Likes
4,514
Somebody needs to send in the r300 to compare and see if the changes were so significant as they say.

They are. The midrange voicing is neutral in R3, and recessed in R300.

There are difference in cabinet construction and damping, not just the baffle, the reflex tubes are heavily damped and the (bass) drivers look different too:

The coax is different too if the cutaways are accurate. Reference has a neo magnet coax, R has a ceramic magnet coax.
 

stren

Active Member
Forum Donor
Joined
Jan 9, 2020
Messages
253
Likes
298
Ref is also slightly larger and a good bit heavier I believe. But yes, different drivers, and presumably more complex crossover etc. Everything is similar but different. Much like the Revel f208 vs the f228be though the price difference there is only 2x ;) You also have to bear in mind too that the reference came a few years earlier. Like most tech, the innovations start in the high end and trickle down over time. If you wait long enough you usually get 90-110% of the performance for a much lower price ;)
 

MZKM

Major Contributor
Forum Donor
Joined
Dec 1, 2018
Messages
4,240
Likes
11,462
Location
Land O’ Lakes, FL
Was browsing Reddit and saw a user post their impressions of the R3 with Dirac, they included the measurement in Dirac’s program, and it has a 1kHz dip:

oNVp8tq.png


Now, the graph that @amirm got doesn’t show this; however, as mentioned before, the graphs that @napilopez got does show this:

R3-2.png


Hi-Fi News also shows it:

1218kef.response.jpg


But also doesn’t show up in this graph:

index.php


No difference in horizontal/vertical positioning looks to be able to cause this, weird.

Maybe KEF did some tweaks and the ones showing the dip are from the pre-fix batches? If whoever lent Amir the R3 measured his, that would be interesting.
 
Last edited:

aarons915

Addicted to Fun and Learning
Forum Donor
Joined
Oct 20, 2019
Messages
685
Likes
1,140
Location
Chicago, IL
Maybe KEF did some tweaks and the ones showing the dip are from the pre-fix batches? If whoever lent Amir the R3 measured his, that would be interesting.

I took a pair home from a local dealer last Thanksgiving and my in room measurements closely match the ones with a 1k dip as well. My room is small so it usually matches pretty closely with a speakers listening window. Ignore below 200Hz, I have some large dips and a crossover in place.

KEFR3.jpg
 

QMuse

Major Contributor
Joined
Feb 20, 2020
Messages
3,124
Likes
2,785
Was browsing Reddit and saw a user post their impressions of the R3 with Dirac, they included the measurement in Dirac’s program, and it has a 1kHz dip:

oNVp8tq.png


Now, the graph that @amirm got doesn’t show this; however, as mentioned before, the graphs that @napilopez got does show this:

R3-2.png


Hi-Fi News also shows it:

1218kef.response.jpg


But also doesn’t show up in this graph:

index.php


No difference in horizontal/vertical positioning looks to be able to cause this, weird.

Maybe KEF did some tweaks and the ones showing the dip are from the pre-fix batches? If whoever lent Amir the R3 measured his, that would be interesting.

What is also present on all these measruements is peak at app 9kHz. Again not shown on that small graph.
 

napilopez

Major Contributor
Forum Donor
Joined
Oct 17, 2018
Messages
2,109
Likes
8,420
Location
NYC
Was browsing Reddit and saw a user post their impressions of the R3 with Dirac, they included the measurement in Dirac’s program, and it has a 1kHz dip:

oNVp8tq.png


Now, the graph that @amirm got doesn’t show this; however, as mentioned before, the graphs that @napilopez got does show this:

R3-2.png


Hi-Fi News also shows it:

1218kef.response.jpg


But also doesn’t show up in this graph:

index.php


No difference in horizontal/vertical positioning looks to be able to cause this, weird.

Maybe KEF did some tweaks and the ones showing the dip are from the pre-fix batches? If whoever lent Amir the R3 measured his, that would be interesting.

Haha this is a small issue that somehow still bothers me every now and then about the R3

I thought it might've been distance related when I initially performed the measurement, but it also shows up in some of my in-room measurements and at 2m test I'd done. It's also in the R11's measurements over at SoundStage network, which makes me think it's a specific issue with the R-series design:
Snag_4e8b299.png


Moreover, it's even in KEF's measurements. The R3 doesn't show it much, but there's a little blip around 900Hz:
Snag_50c7ae0.png

Meanwhile, it's very apparent in the R7 and R11's measurements:
Snag_50cd0d0.png

The only thing I can think of is that one complaint people had about the R3 is that the 'shadow flare' waveguide extension. was sometimes unevely set on the baffle. It's supposed to stick out a little so the Uni-Q unit doesn't 'see' the edge of the cabinet. But I'm not sure if that would have an effect as far down as 1kHz.

I also wondered if maybe there's a reflection happening against the trim ring of the woofer, which juts out a few mm from the cabinet.

Not sure if this is helpful but looking through my mess of R3 measurements (I was much less organized then, unfortunately), I do have a summation of the port and woofer. I'd done the scaling/summation wrong and for some reason didn't save my nearfield woofer measurements to fix it, which is why I didn't publish the bass splice with my measurements. But looking at what I do have, there's clearly something going on at 1K:

R3 Mystery.jpg

It'd be interesting to know whether there was aa silent revision or if this is just a QC issue. I thought the mids were teensy recessed on the R3 and this is probably why.
 

Pepperjack

Member
Joined
May 9, 2020
Messages
90
Likes
64
Hmmm so I am wondering what you guys think. I have been comparing the r5 and focal aria 926. I finally learned that I could do a-b comparisons easily with my Denon though can’t use any amplifier with it when doing so.

when doing this it became extremely obvious that the aria has everything the r5 has but is much more open sounding and clear. The r5 sounds muffled in comparison. In some songs this does allow the singer to seem really intimate and close when closing eyes, but in general everything is more mushy and sounds more muffled on the r5 even with different postitions and toe in etc. Was curious if any data in the r3 would explain this?
 

McFly

Addicted to Fun and Learning
Joined
Mar 12, 2019
Messages
903
Likes
1,861
Location
NZ
Haha this is a small issue that somehow still bothers me every now and then about the R3

I thought it might've been distance related when I initially performed the measurement, but it also shows up in some of my in-room measurements and at 2m test I'd done. It's also in the R11's measurements over at SoundStage network, which makes me think it's a specific issue with the R-series design:

..............................................................................................

It'd be interesting to know whether there was aa silent revision or if this is just a QC issue. I thought the mids were teensy recessed on the R3 and this is probably why.

Chuck the port plugs in - either the outer O or the full plug - and remeasure. Bet you the 'mystery' will be solved.

Ill do some measurements ... when I get mine back ..... 4 months now - to be fair I'm not pushing the supplier due to the virus.
 

HammerSandwich

Major Contributor
Forum Donor
Joined
Nov 22, 2018
Messages
1,137
Likes
1,497
I have been comparing the r5 and focal aria 926. I finally learned that I could do a-b comparisons easily with my Denon though can’t use any amplifier with it when doing so.
How do you match their SPLs? If not at all, that's a big factor.

...it became extremely obvious that the aria has everything the r5 has but is much more open sounding and clear. The r5 sounds muffled in comparison.
I don't recall measurements for either speaker, but Soundstage & Stereophile have measured similar KEF R & Aria 9x6 models. Wild-ass speculation alert!

If the 926 follows its siblings with a higher impedance across the top ~2 octaves, that could add a little output up there. In addition, the Focal almost certainly has more off-axis output above 2kHz or so. Combine the 2 factors, and you may simply be hearing less treble from the KEF.

It will be harder to compare, but try adding a small treble boost to the KEF. Does that reduce the difference? Also, listen to only the KEFs for a while. No EQ, and not 1-2 tracks but an hour or more. Then try one of your "Ah! Too bright!" recordings. Immediately repeat the same track on the Arias. Still airy or are they piercing? Piercing could be accurate, of course, but this should help spot a treble difference.
 

Pepperjack

Member
Joined
May 9, 2020
Messages
90
Likes
64
This is what can happen when you compare speakers that have bass ports in different places--Focal on the front and KEF on the back. Once you figure out how to deal with the room mode the KEFs seem to be activating (by using a plug as mentioned, by moving it away from the wall, or through DSP), it'll sound like what you expecting it to.

I will give it a shot but have used them for a couple of weeks and tried various positions, they seemed best where they are now actually, but I’ll run a test with port plugs again tomorrow.
 

Pepperjack

Member
Joined
May 9, 2020
Messages
90
Likes
64
How do you match their SPLs? If not at all, that's a big factor.


I don't recall measurements for either speaker, but Soundstage & Stereophile have measured similar KEF R & Aria 9x6 models. Wild-ass speculation alert!

If the 926 follows its siblings with a higher impedance across the top ~2 octaves, that could add a little output up there. In addition, the Focal almost certainly has more off-axis output above 2kHz or so. Combine the 2 factors, and you may simply be hearing less treble from the KEF.

It will be harder to compare, but try adding a small treble boost to the KEF. Does that reduce the difference? Also, listen to only the KEFs for a while. No EQ, and not 1-2 tracks but an hour or more. Then try one of your "Ah! Too bright!" recordings. Immediately repeat the same track on the Arias. Still airy or are they piercing? Piercing could be accurate, of course, but this should help spot a treble difference.

I am pretty new to all of this but listened to the kefs exclusively for a couple of weeks and was able to keep them in what I thought was their optimum position. I liked the 926 (listened to for a couple weeks now also) a bit farther apart. Was very happy with the kefs (they stomped my Unifi in pretty much every way, such that it was impossible to do a blind comparison, since no matter how I positioned them etc I always knew as soon as the Unifi came on) but when comparing a-b like this it seems the aria is winning everything hands. Voices are clearer, instruments are clearer, Bass seems better. It’s like the soundstage for the Kef drops down vertically and shrinks in, toward the center, and everything is sort of more compressed across the every song and the full spectrum. Enough that I started playing with toe in again but it never was able to really come close. So, yea, I have been surprised. If I just listened to the Kef alone I never would have thought I was missing anything.

interestingly when blind testing I had one song for each speaker I though sounded a bit piercing and metallic. When I listened to entire songs by each speaker and a/b blind I mostly preferred the focal but sometimes the Kef, but the quick change a/b is no competition at the moment. Will give the plugs tomorrow and see If it can bring it to life for me.

Edit: I forgot to mention. To set the spl I first tried to go by some pink noise and a imm6 mic on audio tools. Then I followed up with a test tone on each individual speaker from listening position to determine that I need to move up the volume two notches for the Kef. The difference was so stark though that I was worried that it was still an spl issue so I started cranking up the volume until the Kef was obviously a bit louder (4-5 clicks, sometimes as high as 8 more), yet, despite being louder still seemed more “muffled” and constricted across the sound stage in voices and instruments etc. so it’s definitely not an spl issue. will try hooking it back up to the amplifier tomorrow and see if that Helps the Kef but I can’t do as easy or quick an a/b that way.

I will swap their physical positions tomorrow, even though I chose these spots for each individually, I suppose I could have just chosen wrong for the Kef.
 
Last edited:

Pepperjack

Member
Joined
May 9, 2020
Messages
90
Likes
64
well granted it’s hasn’t been long but I immediately did go ahead and swap their positions, and it was the exact same results. On a couple of classical pieces the Kef was closer, but on every song the a/b was extremely obvious and easily a win for the aria imo. Did a couple of songs blind and was easily able to spot which was which due to the differences described above, changing the positions didn’t really seem to take anything from the aria either though the Kef seemed a slight improvement. Dont have the port plugs on hand so will check the plugs tomorrow but, if I’m not crazy, the arias just pretty easily stomped the r5 in my room.
 

muad

Senior Member
Joined
Jun 8, 2019
Messages
415
Likes
471
well granted it’s hasn’t been long but I immediately did go ahead and swap their positions, and it was the exact same results. On a couple of classical pieces the Kef was closer, but on every song the a/b was extremely obvious and easily a win for the aria imo. Did a couple of songs blind and was easily able to spot which was which due to the differences described above, changing the positions didn’t really seem to take anything from the aria either though the Kef seemed a slight improvement. Dont have the port plugs on hand so will check the plugs tomorrow but, if I’m not crazy, the arias just pretty easily stomped the r5 in my room.

Interesting. Too bad you don't have a measurement mic. Would be nice to compare their in room measurements from your listening couch.

I have a custom 3way that in room measures mostly flat to 4 khz and then tapers off. When I eq it to fit the Harman curve it sounds much less clear, but better balanced. I guess I prefer it muffled, go figure :)
 

Pepperjack

Member
Joined
May 9, 2020
Messages
90
Likes
64
Interesting. Too bad you don't have a measurement mic. Would be nice to compare their in room measurements from your listening couch.

I have a custom 3way that in room measures mostly flat to 4 khz and then tapers off. When I eq it to fit the Harman curve it sounds much less clear, but better balanced. I guess I prefer it muffled, go figure :)

I doubt what I am hearing as muffled is the same thing happening in yours. This is a pretty stark difference though if I hadn’t heard the aria a/b I would have been happy with it and never thought it was possible. I suspect it is either something about the less than ideal room or perhaps a bad sample from Kef. but now that I have heard the difference going back is not really an option.

Been thinking about a better mic but haven’t committed to it yet. Knowing me, I’ll probably order something tomorrow :)
 
Top Bottom