• 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

richard12511

Major Contributor
Forum Donor
Joined
Jan 23, 2020
Messages
4,335
Likes
6,700
That's... A long video:). I'll watch eventually, but Any key points?

- The examples they used for good neutral narrow and good neutral wide dispersion speakers were the JTR Noesis 212 and the BMR, respectively. I really like how they normalized the polar maps. That made it much easier to see the differences in dispersion width. BMR was -6db at 90 degrees. 212 was -6db at 30 degrees. I wish reviews would normalize these graphs as they did here.
- Narrow dispersion gives a wider sweetspot when toed in in front of listener. I'm not sure I agree with this.
- Narrow dispersion has the same effect as absorbing the first reflections, only it does it much better. The problem with absorbing first reflections is that absorbers do a very poor job of preserving the tonal balance. Absorbers don't absorb lower frequencies well, and can often be reflective at higher frequencies, essentially causing a midrange dip.

* Not enough real research to support the points below, but most experts seem to agree, anecdotally
- Narrow dispersion will have a tighter more focused image
- Wide dispersion will have a larger and wider soundstage, with more space between instruments
- Narrow dispersion probably preferable for recordings that were recorded in absorptive rooms (Pop, electronic?)
- Wide dispersion probably preferable for recordings that were recorded in real rooms(live music, orchestral, acoustic?)
- Narrow dispersion maybe preferable for surround systems. <- More research needed
- Wide dispersion maybe preferable for 2 channel. <- More research needed
- Mono tests possibly insufficient for teasing out preferences of wide vs narrow. Only dispersion element present in mono is spaciousness. <- More research needed. He gives an example of good speakers that could be used to test. The example given for perhaps the best wide and narrow designs he's seen are Vivid Audio and D&D 8C, respectively.

Overall main point seems to be that insufficient research has been done to really say which is better, and even if the research is done, it may depend on room, listening distance, and source material. I thought it did a good job of highlighting the advantages of narrow and wide dispersion.
 
Last edited:

napilopez

Major Contributor
Forum Donor
Joined
Oct 17, 2018
Messages
2,109
Likes
8,420
Location
NYC
- The examples they used for good neutral narrow and good neutral wide dispersion speakers were the JTR Noesis 212 and the BMR. I like how they normalized the polar maps. That made it much easier to see the differences in dispersion width. BMR was -6db at 90 degrees. 212 was -6db at 30 degrees. I wish reviews would normalize these graphs as they did here
- Narrow dispersion gives a wider sweetspot when toed in in front of listener. I'm not sure I agree with this.
- Narrow dispersion has the same as effect as absorbing the first reflections, only it does it much better. The problem with absorbing first reflections is that absorbers do a very poor job of preserving the tonal balance. Absorbers don't absorb lower frequencies well, and can often be reflective at higher frequencies, essentially causing a midrange dip.

* Not enough real research to support the points below, but most experts seem to agree, anecdotally
- Narrow dispersion will have a tighter more focused image
- Wide dispersion will have a larger and wider soundstage, with more space between instruments
- Narrow dispersion probably preferable for recordings that were recorded in absorptive rooms (Pop, electronic?)
- Wide dispersion probably preferable for recordings that were recorded in real rooms(live music, orchestral, acoustic?)
- Narrow dispersion maybe preferable for surround systems. <- More research needed
- Wide dispersion maybe preferable for 2 channel. <- More research needed
- Mono tests possibly insufficient for teasing out preferences of wide vs narrow. Only dispersion element present in mono is spaciousness. <- More research needed. He gives an example of good speakers that could be used to test. The example given for perhaps the best wide and narrow designs he's seen are Vivid Audio and D&D 8C, respectively.

Overall main point seems to be that insufficient research has been done to really say which is better, and even if the research is done, it may depend on room, listening distance, and source material. I thought it did a good job of highlighting the advantages of narrow dispersion. It did highlight the advantages of wide dispersion, but didn't do as good of a job(imo).

Thank you for the summary! Looks like a well done video. I always normalize my polar graphs for that reason. Imo they aren't very useful, especially for comparison, without normalization.

I also agree with narrow directivity crossed in front having a wider sweetspot. This makes sense and I've usually found it to be true. Not sure if you're aware or if it's explained in the video, but the idea is if you cross the speakers a little bit in front of your listening position, it makes it such that if you move to the left, closer to the left speaker, you're also moving more on axis to the right speaker. So with ideal positioning, the SPL balance remains the same between the two speakers even as you move within a large sweetspot. Though I don't see why you can do this with narrower speakers too.

In practice though, I've found it difficult to find the right balance of wide sweetspot and good imaging in my home. I also find it aesthetically objectionable, thought it's not as bad if your speakers are in the corner or in an office setup.

The one point I'm not sure I agree with is narrow being better for absorptive room recordings and wide for real rooms. I'd kind of think the opposite because absorptive rooms have less room information built in so combined with fewer sidewall reflections you just end up with a very dry sound... But I don't have strong opinion on this.

I'll be sure to watch at some point, seems like good content.
 
Last edited:

andreasmaaan

Master Contributor
Forum Donor
Joined
Jun 19, 2018
Messages
6,652
Likes
9,399
Great summary @richard12511, thanks.

Narrow dispersion has the same as effect as absorbing the first reflections, only it does it much better. The problem with absorbing first reflections is that absorbers do a very poor job of preserving the tonal balance. Absorbers don't absorb lower frequencies well, and can often be reflective at higher frequencies, essentially causing a midrange dip.

This is an interesting point. I appreciate where they're coming from, but OTOH, putting aside dipoles, cardioids and massive horns, speakers are essentially omni below a few hundred Hertz, and often a bit higher. From this perspective, side/front wall absorption may actually be a more effective solution if reflections from those points are undesired.

For example, it's quite difficult to find a speaker that is highly directional at say 500Hz, but it is not that hard to build/buy a panel that will absorb reflections at this frequency.
 

VintageFlanker

Major Contributor
Forum Donor
Joined
Sep 20, 2018
Messages
4,940
Likes
19,691
Location
Paris
Capture d’écran (39).png


"No research has been conducted to understand subjective preferences of directivity"

Is that correct, @Matthew J Poes ?
 

andreasmaaan

Master Contributor
Forum Donor
Joined
Jun 19, 2018
Messages
6,652
Likes
9,399
"No research has been conducted to understand subjective preferences of directivity"

It's basically wrong. I can name a number of studies.

A large number (but not all) of these are summarised here.

It would be more correct to say that, although research has been conducted, the findings are ambiguous.

(Sorry, I do realise I'm not @Matthew J Poes)
 

richard12511

Major Contributor
Forum Donor
Joined
Jan 23, 2020
Messages
4,335
Likes
6,700
Thank you for the summary! Looks like a well done video. I always normalize my polar graphs for that reason. Imo they aren't very useful, especially for comparison, without normalization.

I also agree with narrow directivity crossed in front having a wider sweetspot. This makes sense and I've usually found it to be true. Not sure if you're aware or if it's explained in the video, but the idea is if you cross the speakers a little bit in front of your listening position, it makes it such that if you move to the left, closer to the left speaker, you're also moving more on axis to the right speaker. So with ideal positioning, the SPL balance remains the same between the two speakers even as you move within a large sweetspot. Though I don't see why you can do this with narrower speakers too.

They do a good job of explaining it in the video. I agree with his and your point, and it's the way I have my JTRs setup in my living room. I guess the part I disagree with is calling that an advantage of narrow dispersion. I see it as both an advantage and disadvantage, as not everyone can toe their speakers in that much. As you said, it's quite objectionable, aesthetically. With no toe in, for example, I find that the sweet spot is actually wider with wide dispersion.


The one point I'm not sure I agree with is narrow being better for absorptive room recordings and wide for real rooms. I'd kind of think the opposite because absorptive rooms have less room information built in so combined with fewer sidewall reflections you just end up with a very dry sound... But I don't have strong opinion on this.

I'll be sure to watch at some point, seems like good content.

That was my first thought, too. My thought was that if the spaciousness is already in the recording, I don't want to add more. I think he was coming at it from the point of ~"the closer you are to what the mixing engineer was hearing when designing the song, the better the sound will be". He's a member here, so maybe he can better explain his view. One example they gave was a really nice MBL that they heard. They thought it sounded wonderful for classical music, but quite unnatural for rock music. Anecdotally, when listening to my JTRs, Classical Music(as long as there's no solo vocals) is the only music I prefer to listen to upmixed to "Hall in Munich". It essentially adds more reflective sound, so it's probably similar to increasing dispersion width. Maybe I do agree with him? I'm really not sure, and like you, I don't have a strong opinion one way or the other.
 
Last edited:

richard12511

Major Contributor
Forum Donor
Joined
Jan 23, 2020
Messages
4,335
Likes
6,700
This is an interesting point. I appreciate where they're coming from, but OTOH, putting aside dipoles, cardioids and massive horns, speakers are essentially omni below a few hundred Hertz, and often a bit higher. From this perspective, side/front wall absorption may actually be a more effective solution if reflections from those points are undesired.

I agree with your point here. There's probably a frequency range for which panels are more effective(say 100Hz to 500Hz?). That said, I don't think it matters for the comparison they used here (JTR Noesis 212 vs Philharmonic BMR), as the JTR only controls directivity down to 500Hz, and looking at the polar maps, the two speakers have very, very similar dispersion below 500Hz. So yeah, panels are no doubt more effective at controlling dispersion below 500Hz than the JTR, but perhaps a better test would be panels vs something that controls down even lower, like the D&D 8C.
 
Last edited:

Jon AA

Senior Member
Forum Donor
Joined
Feb 5, 2020
Messages
465
Likes
905
Location
Seattle Area
I also agree with narrow directivity crossed in front having a wider sweetspot. This makes sense and I've usually found it to be true. Not sure if you're aware or if it's explained in the video,
He does explain it a little, but the third upcoming video will apparently be dedicated to talking about this subject a lot.
but the idea is if you cross the speakers a little bit in front of your listening position, it makes it such that if you move to the left, closer to the left speaker, you're also moving more on axis to the right speaker. So with ideal positioning, the SPL balance remains the same between the two speakers even as you move within a large sweetspot. Though I don't see why you can do this with narrower (wider?) speakers too.
Did you mean wider above? If so, I think it simply comes down to not getting enough volume reduction off axis with any reasonable amount of toe. Taking the BMR's for example, you'd need to toe the speakers in so much they're basically facing one another to get a meaningful SPL reduction of the close speaker as you move off axis. Obviously that probably wouldn't sound very good.

I've been a fan of this concept for a while, not just for the sweetspot size in stereo listening, but particularly for the left and right channels in a multichannel setup--where concern for multiple listeners is much more common. For a listener sitting toward the left front area of a home theater, with really wide dispersion speakers that left speaker will tend to drown out the center channel (and everything else) to a much greater extent than a more narrow constant directivity design.

Something Matthew didn't specifically point out, which I hope he does in the next video, is that it's not just a "narrow/wide" thing. It's always going to work better with more constant directivity designs. Otherwise, as you get off axis, the volume might be reduced, but the tonal balance will change significantly. As an example, if you're sitting 60 degrees off axis from a HDI-1600:

HDI1600HorizOffAxis.jpg


You can see the volume is much reduced, but you're still getting a pretty flat response. Compare that with an even more narrow RP-600M:

RP600HorizOffAxis.jpg


As you can see, the tonal balance will be radically different (treble completely disappearing) for the off-axis speaker compared with the farther, on-axis speaker which isn't going to sound nearly as good. In this case, the "wider dispersion" speaker pair will be the better performer for such a setup because it's more nearly constant directivity. Though since it's a bit wider it may need a bit more toe to get the effect. The above assumes both are EQ'd flat on-axis, of course.

I look forward to the third video, hopefully he brings this up. ;) Back to the above video, I think it was well done and about the only thing I think I disagree with is where the "arbitrary line" was drawn between "narrow" and "wide" being shifted toward the wide end of the spectrum a bit too far. He had the JBL M2 as an example of a narrow speaker...and in my opinion it's pretty darn wide (or at least in the "middle of the spectrum"). There aren't a whole lot of speakers as wide as the BMR that measure smooth off-axis. Though I guess there are plenty that are that wide somewhere in the frequency response, but to me they don't count when you're talking about good speakers. It's an arbitrary line though, so I guess any disagreement is semantics. ;)
 
Last edited:

napilopez

Major Contributor
Forum Donor
Joined
Oct 17, 2018
Messages
2,109
Likes
8,420
Location
NYC
He does explain it a little, but the third upcoming video will apparently be dedicated to talking about this subject a lot.

Did you mean wider above? If so, I think it simply comes down to not getting enough volume reduction off axis with any reasonable amount of toe. Taking the BMR's for example, you'd need to toe the speakers in so much they're basically facing one another to get a meaningful SPL reduction of the close speaker as you move off axis. Obviously that probably wouldn't sound very good.

I've been a fan of this concept for a while, not just for the sweetspot size in stereo listening, but particularly for the left and right channels in a multichannel setup--where concern for multiple listeners is much more common. For a listener sitting toward the left front area of a home theater, with really wide dispersion speakers that left speaker will tend to drown out the center channel (and everything else) to a much greater extent than a more narrow constant directivity design.

Something Matthew didn't specifically point out, which I hope he does in the next video, is that it's not just a "narrow/wide" thing. It's always going to work better with more constant directivity designs. Otherwise, as you get off axis, the volume might be reduced, but the tonal balance will change significantly. As an example, if you're sitting 60 degrees off axis from a HDI-1600:

View attachment 65615

You can see the volume is much reduced, but you're still getting a pretty flat response. Compare that with an even more narrow RP-600M:

View attachment 65616

As you can see, the tonal balance will be radically different (treble completely disappearing) for the off-axis speaker compared with the farther, on-axis speaker which isn't going to sound nearly as good. In this case, the "wider dispersion" speaker pair will be the better performer for such a setup because it's more nearly constant directivity. Though since it's a bit wider it may need a bit more toe to get the effect. The above assumes both are EQ'd flat on-axis, of course.

I look forward to the third video, hopefully he brings this up. ;) Back to the above video, I think it was well done and about the only thing I think I disagree with is where the "arbitrary line" was drawn between "narrow" and "wide" being shifted toward the wide end of the spectrum a bit too far. He had the JBL M2 as an example of a narrow speaker...and in my opinion it's pretty darn wide (or at least in the "middle of the spectrum"). There aren't a whole lot of speakers as wide as the BMR that measure smooth off-axis. Though I guess there are plenty that are that wide somewhere in the frequency response, but to me they don't count when you're talking about good speakers. It's an arbitrary line though, so I guess any disagreement is semantics. ;)

I did mean wider, and you're totally right about everything else. Though keep in mind 'extreme' toe in won't be that much off axis. For example, toeing in speakers so they cross about a foot in front of my listening position only actually puts me about 10-15 degrees off axis. It's just off axis in the other direction.
 

Jon AA

Senior Member
Forum Donor
Joined
Feb 5, 2020
Messages
465
Likes
905
Location
Seattle Area
Certainly, the angle you want to look at will depend on setup and what seat you're evaluating. 60 degrees is pretty extreme. For example, in a typical equilateral triangle setup, with the speakers toed in to face the sweet spot exactly, somebody sitting directly in front of the left speaker will only be 30 degrees off axis from it. With 10-20 degrees of "extreme" toe in that'll be 40-50 degrees. You'll get to 60 if you look at one seat farther to the left or have more toe. One would need to look at their available seating locations and speaker layout on a case by case basis to see what's going on.
 

richard12511

Major Contributor
Forum Donor
Joined
Jan 23, 2020
Messages
4,335
Likes
6,700
He does explain it a little, but the third upcoming video will apparently be dedicated to talking about this subject a lot.

Did you mean wider above? If so, I think it simply comes down to not getting enough volume reduction off axis with any reasonable amount of toe. Taking the BMR's for example, you'd need to toe the speakers in so much they're basically facing one another to get a meaningful SPL reduction of the close speaker as you move off axis. Obviously that probably wouldn't sound very good.

I've been a fan of this concept for a while, not just for the sweetspot size in stereo listening, but particularly for the left and right channels in a multichannel setup--where concern for multiple listeners is much more common. For a listener sitting toward the left front area of a home theater, with really wide dispersion speakers that left speaker will tend to drown out the center channel (and everything else) to a much greater extent than a more narrow constant directivity design.

Something Matthew didn't specifically point out, which I hope he does in the next video, is that it's not just a "narrow/wide" thing. It's always going to work better with more constant directivity designs. Otherwise, as you get off axis, the volume might be reduced, but the tonal balance will change significantly. As an example, if you're sitting 60 degrees off axis from a HDI-1600:

View attachment 65615

You can see the volume is much reduced, but you're still getting a pretty flat response. Compare that with an even more narrow RP-600M:

View attachment 65616

As you can see, the tonal balance will be radically different (treble completely disappearing) for the off-axis speaker compared with the farther, on-axis speaker which isn't going to sound nearly as good. In this case, the "wider dispersion" speaker pair will be the better performer for such a setup because it's more nearly constant directivity. Though since it's a bit wider it may need a bit more toe to get the effect. The above assumes both are EQ'd flat on-axis, of course.

I look forward to the third video, hopefully he brings this up. ;) Back to the above video, I think it was well done and about the only thing I think I disagree with is where the "arbitrary line" was drawn between "narrow" and "wide" being shifted toward the wide end of the spectrum a bit too far. He had the JBL M2 as an example of a narrow speaker...and in my opinion it's pretty darn wide (or at least in the "middle of the spectrum"). There aren't a whole lot of speakers as wide as the BMR that measure smooth off-axis. Though I guess there are plenty that are that wide somewhere in the frequency response, but to me they don't count when you're talking about good speakers. It's an arbitrary line though, so I guess any disagreement is semantics. ;)

I definitely with you that the M2 should be considered a wide dispersion loudspeaker. It's not as wide as the Salon 2 or BMR, but it has a 120 x 100 degree horn. It's a wide dispersion design.
 
  • Like
Reactions: 617

stren

Active Member
Forum Donor
Joined
Jan 9, 2020
Messages
253
Likes
298
So for home theatre use - is narrow dispersion preferred? Presumably the higher the speaker count - the more you would want a narrow dispersion?
 

richard12511

Major Contributor
Forum Donor
Joined
Jan 23, 2020
Messages
4,335
Likes
6,700
So for home theatre use - is narrow dispersion preferred? Presumably the higher the speaker count - the more you would want a narrow dispersion?

From what I understand: Most people seem to anecdotally agree with this, but there's very little (if any) science to support this.
 

napilopez

Major Contributor
Forum Donor
Joined
Oct 17, 2018
Messages
2,109
Likes
8,420
Location
NYC
So for home theatre use - is narrow dispersion preferred? Presumably the higher the speaker count - the more you would want a narrow dispersion?

That makes sense to me. It's always a balancing act, but if you could theoretically have a 1,000 channel system, you'd want each of those channels to be focused accurately rather than dispersing sound willy billy. Conversely, if you're only listening to a mono speaker, you want that to have as wide directivity as possible to not reveal itself as a small singular source.
 

andreasmaaan

Master Contributor
Forum Donor
Joined
Jun 19, 2018
Messages
6,652
Likes
9,399
So yeah, panels are no doubt more effective at controlling dispersion below 500Hz than the JTR, but perhaps a better test would be panels vs something that controls down even lower, like the D&D 8C.

For sure. The 8C would absolutely be superior to panels in this regard, I would think.
 

Dmi3i

Member
Joined
May 12, 2020
Messages
8
Likes
8
but what if at the recording studio the person who mixed the composition that Amirm listened to in headphones, or monitors with such an AFC, in which this composition sounds on R3 just like that ... and on m16 is more emotional) If you do it with an equalizer, the AFC R3 = m16 (how real is it) Will there be a result?) and then vice versa m16 = r3 ...
 

Haint

Senior Member
Joined
Jan 26, 2020
Messages
346
Likes
452
From what I understand: Most people seem to anecdotally agree with this, but there's very little (if any) science to support this.

There is an argument for the front 3, but side and/or rears very often act as effective mono sources which is why dipoles are so popular for those channels. Then in mixing and matching for directivity, you would miss out on the theoretical ideal of 5 or 7+ identical speakers.
 

stren

Active Member
Forum Donor
Joined
Jan 9, 2020
Messages
253
Likes
298
BTW anyone know if there is a R2C in the review pipe? Be interesting to see how that compares to the R3 and therefore presumably draw some conclusions to the R5 which uses the same smaller bass drivers and smaller baffle around the uni-q.
 

aarons915

Addicted to Fun and Learning
Forum Donor
Joined
Oct 20, 2019
Messages
685
Likes
1,140
Location
Chicago, IL
BTW anyone know if there is a R2C in the review pipe? Be interesting to see how that compares to the R3 and therefore presumably draw some conclusions to the R5 which uses the same smaller bass drivers and smaller baffle around the uni-q.

I think they would also be interesting for the front 3 channels as well for those who may want a sealed speaker or just dual woofers in an MTM setup. They won't dig as deep as an R3 but they should have more output above 100Hz or so.
 
Top Bottom