KEF R3 Speaker Review

[eyes-on-you]

i have a question about the grill for kef owners..
With the grill i feel the mids-highs are a bit messy, with grill off i feel the speaker sounding crearly better, its only me? i saw some measurements from R2C ( Erin ) and the grill do a bit weird work

Hi,

I think me too. Always grills off

 

[alitomr1979]

This is my graph taken with REW. No Sub:

Why the big drop after 10k ???
This is my measurement with my SB1000 sub:

Practically no change.

I suppose its impossible to like a pair of speakers with such a massive peak in the lows, right?

Im going to take manual measures in Roon, because it is where I play 99% of my music.

Thoughts on the graphs?

 

[alitomr1979]

Could you sheare here when the measurements done?

I’m find my R7 is dark too but keep in mind that, a more colorful speaker between 2-5 khz may be coincidentally covering up problems in your room.

What would you like to know? I am measuring in my bedroom. It is about 4m x 3.8m x 2.85m. One small curtains, concrete walls. My bed is a queen size and most walls are uncovered... not damped at all...

 

[BYRTT]

This is my graph taken with REW. No Sub..... .....Why the big drop after 10k ???.....

In general what is called "in room response" or "PIR (predicted in room)" or "estimated in room" is a declining curve as frequency increase, it falls about 0,7dB per octave in farfield, so a target curve for farfield PIR measurements are not a flat curve but a declining one about as seen i below graphs called ideals, drop after 10kHz can be many things as carpet/curtain/long distances/missing calibration file or if REW sample rate setting is mismatched to the operating system or hardware.

.....I suppose its impossible to like a pair of speakers with such a massive peak in the lows, right?.....

Below 500Hz can sometimes look scary but also it can sometimes sound wonderfull even it looks scary, some users pay alot of attention and move around their systems and/or use multiple subs to get smoother curves.

.....Thoughts on the graphs?

As hinted yesterday acoustic PIR measurements are not a walk in the park and need some care and practice to be used as basement for correction, and yours look a bit to rough if you look below the upper one was your phone and the lower one was using REW and they supposed to allign with the orange PIR curve, now take a look at right side and see @thewas PIR curve measurement for IN-8..

 

[BrokenEnglishGuy]

In general what is called "in room response" or "PIR (predicted in room)" or "estimated in room" is a declining curve as frequency increase, it falls about 0,7dB per octave in farfield, so a target curve for farfield PIR measurements are not a flat curve but a declining one about as seen i below graphs called ideals, drop after 10kHz can be many things as carpet/curtain/long distances/missing calibration file or if REW sample rate setting is mismatched to the operating system or hardware.

Below 500Hz can sometimes look scary but also it can sometimes sound wonderfull even it looks scary, some users pay alot of attention and move around their systems and/or use multiple subs to get smoother curves.

As hinted yesterday acoustic PIR measurements are not a walk in the park and need some care and practice to be used as basement for correction, and yours look a bit to rough if you look below the upper one was your phone and the lower one was using REW and they supposed to allign with the orange PIR curve, now take a look at right side and see @thewas PIR curve measurement for IN-8..

View attachment 148169

The dip at 4khz~~~ is about 5dB also the 500hz another dip

 

[Sancus]

drop after 10kHz can be many things as carpet/curtain/long distances/missing calibration file or if REW sample rate setting is mismatched to the operating system or hardware.

If this was measured with a UMIK, those are prone to weirdness above 15khz even with the calibration file.

 

[BYRTT]

If this was measured with a UMIK, those are prone to weirdness above 15khz even with the calibration file.

Had me a UMIK for many years but soon after stepped up to a analog German iSEMcon EMX-7150 so as to get dual channel sync, but what i remember was that its locked at 48kHz so if operating system and soundcard output or REW runs at 44,1kHz then we get sample rate mitchmatch and remember if one care adjust so anything runs at 48kHz UMIK was not bad but miss dual channel sync.

 

[DACs_Lover]

Should we aim the KEF co-axial driver at us directly? From the color chart, if we put the speakers pararell to the front wall we'll lose some of the treble?

Thank you.

 

[Eetu]

Should we aim the KEF co-axial driver at us directly? From the color chart, if we put the speakers pararell to the front wall we'll lose some of the treble?

Thank you.

15 deg off axis seems just about right (see post #8). Parallel to wall (30 deg assuming an equilateral triangle listening setup) may sound a bit dark, depends on your room as well.

 

[alitomr1979]

I am still EQing but even at this early point in time I can say these are impressive speakers. I think I am being too agressive on the some EQ because certain songs sound a little weird.

Thank you @BrokenEnglishGuy and @BYRTT for the support and @amirm for the great content.

Thoughts on how to proceed (with EQ) to get the speaker to sound as flat as possible in my room? No treatment possible for now in this room.

 

[BrokenEnglishGuy]

Should we aim the KEF co-axial driver at us directly? From the color chart, if we put the speakers pararell to the front wall we'll lose some of the treble?

Thank you.

im listening my r7 at the top woofer level, i don't know what angle is but is not that terrible if i put my head in the middle i don't hear much difference xD

I am still EQing but even at this early point in time I can say these are impressive speakers. I think I am being too agressive on the some EQ because certain songs sound a little weird.

Thank you @BrokenEnglishGuy and @BYRTT for the support and @amirm for the great content.

Thoughts on how to proceed (with EQ) to get the speaker to sound as flat as possible in my room? No treatment possible for now in this room.

Thank you, keep working in the EQ for fix the room, that gonna fix the problem not a new dac or something like that

Some recording are weird, a good speaker should show that weirdness, also the BBC dip can improve the sound for poorly recording, thats a dip between 1khz-3khz, but -1dB or -2dB MAX. Its something like that

 

[NTK]

Should we aim the KEF co-axial driver at us directly? From the color chart, if we put the speakers pararell to the front wall we'll lose some of the treble?

Thank you.

Dr Jack Oclee-Brown of KEF recommends listening at least 10 degrees off-axis.
https://www.audiosciencereview.com/...w-bookshelf-speaker.19216/page-12#post-632572

 

[aarons915]

I am still EQing but even at this early point in time I can say these are impressive speakers. I think I am being too agressive on the some EQ because certain songs sound a little weird.

Thank you @BrokenEnglishGuy and @BYRTT for the support and @amirm for the great content.

Thoughts on how to proceed (with EQ) to get the speaker to sound as flat as possible in my room? No treatment possible for now in this room.

Many of us have had better luck EQ'ing above the transition frequency using the CTA-2034 style measurements because the most important sounds have been found to be the direct sound followed by the early reflections. I did a lot of experimenting with the R3 and this was pretty much the EQ I ended up with that sounded neutral and smooth, the most important thing is to slightly boost the region between 1-2k and EQ the slight ER reflections peak around 2.7K because it makes them sound a tad harsh. Here's the EQ and resulting CTA-2034 curves after EQ. Side note, I use the listening window because it is assumed that they will be listened to around 10-20 degrees off axis, also make sure the shadow flare is seated around the woofer and midrange cone.

 

[phoenixdogfan]

Ok, thanks. I found it. I was trying to apply somefilters in Roon but I cannot do it from the phone. Weird.
I definitely need to run some measurements and start tweaking.

I’ll keep you posted.

Remember, because it was a room mode, it was peculiar to Amir's listening environment. The R3s in your room may excite a mode at a different frequency and intensity, and may require very different correction or none at all.

 

[MaxBuck]

15 deg off axis seems just about right (see post #8). Parallel to wall (30 deg assuming an equilateral triangle listening setup) may sound a bit dark, depends on your room as well.

Wow, thanks! Good confirmation for how I've set the toe-in of my R11s. They sound terrific, even in my horribly suboptimal listening space.

 

[eyes-on-you]

15 deg off axis seems just about right (see post #8). Parallel to wall (30 deg assuming an equilateral triangle listening setup) may sound a bit dark, depends on your room as well.

How it is possible? On axis looks more trable on post 8.

Could you explain?

 

[Eetu]

How it is possible? On axis looks more trable on post 8.

Could you explain?

The R3 is a bit bright on axis (as noted by Amir as well), I pointed out that the treble response is more neutral slightly off axis (~15 deg). This is confirmed by the designer Jack Oclee-Brown (post #1,253).

DACs_Lover also asked about placing the speaker parallel to front wall (aiming straight into the room) and in a normal stereo listening position this means listening 30 degrees off axis which may sound a bit dark (treble fall-off). So 10-20 deg off-axis seems optimal. YMMV since this depends how reflective your room is, how close to side walls the speakers are placed etc.

 

[alitomr1979]

The R3 is a bit bright on axis (as noted by Amir as well), I pointed out that the treble response is more neutral slightly off axis (~15 deg). This is confirmed by the designer Jack Oclee-Brown (post #1,253).

DACs_Lover also asked about placing the speaker parallel to front wall (aiming straight into the room) and in a normal stereo listening position this means listening 30 degrees off axis which may sound a bit dark (treble fall-off). So 10-20 deg off-axis seems optimal. YMMV since this depends how reflective your room is, how close to side walls the speakers are placed etc.

Casually I am listening to all LS50, LS50W and R3 at 10-15 deg of-axis. I settled long ago in that position not to reduce brightness but because it offered a good compromise between losing brightness and maintaining a wide soundstage. I like bright speakers. Go figure!

When I say bright I mean neutral, flat speakers, I think, because flat seems to be bright to a lot of people, as is the case with the R3.
I haven’t got too much time for the EQing but I like what I am hearing. Coming from the LS50 and LS50W the word to describe the R3 is imposing. Yet so detailed. I like them.

 

[dshreter]

I'm looking at KEF R3 and Neumann KH120 for stereo home theater and music. I know there are somewhat different implications on the interface needed for each, but any thoughts on differences in sound quality between the two?

 

[aarons915]

I'm looking at KEF R3 and Neumann KH120 for stereo home theater and music. I know there are somewhat different implications on the interface needed for each, but any thoughts on differences in sound quality between the two?

I had the R3 and tried the KH 120 and really liked them, so much that pretty much immediately decided to box up the R3 and not look back. The problem was after listening to the KH 120 for longer periods in a living room setup they started really fatiguing me worse than any speaker I've ever had, my ears were actually feeling numb. The weird thing is they still sounded great even while fatiguing me. I tried quite a few different things including EQ and couldn't stop the fatigue so I had to send them back. I'm fairly convinced the vertical response is responsible for doing funky things in a living room setup because the direct sound and sidewall reflections are pristine, which is why I think they sounded so great yet were making my ears go numb. So not the best answer but it might be a good idea to buy both and compare them in your room, the R3 are good speakers but do benefit from a bit of EQ as I showed some filters a few posts back. It would be nice to get a 2nd opinion on using the KH 120 in a living room.