KEF R3 meta Measurements

[cavedriver]

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

My point is that there is no indication in ALL the other reviews on this site that narrower dispersion is better when you look at most of the highest scoring speakers. None of those speakers have dispersion this narrow, and in fact Amir often remarks that it is a positive that the dispersion is at least 50 degrees. And yet, here several people are suggesting that this speaker is superlative because it has narrow dispersion. I agree that it is a matter of rooms and preferences, but I do not agree that this speaker is broadly "better" than other speakers and in my experience speakers like this are difficult to position and have narrow sweet spots from which you can enjoy the stereo illusion, making them unsuitable for many listening rooms and environments. I think we all would agree that the dispersion should be "smooth", and if that is what Nuyes meant by "precise directional control" then I would agree, but I feel it should be noted that this speaker's dispersion width runs on the narrow side and that is not necessarily a positive. Frankly, I would not be buying this speaker because narrow sweet spots are such a pita.

 

[BrokenEnglishGuy]

My point is that there is no indication in ALL the other reviews on this site that narrower dispersion is better when you look at most of the highest scoring speakers. None of those speakers have dispersion this narrow, and in fact Amir often remarks that it is a positive that the dispersion is at least 50 degrees. And yet, here several people are suggesting that this speaker is superlative because it has narrow dispersion. I agree that it is a matter of rooms and preferences, but I do not agree that this speaker is broadly "better" than other speakers and in my experience speakers like this are difficult to position and have narrow sweet spots from which you can enjoy the stereo illusion, making them unsuitable for many listening rooms and environments. I think we all would agree that the dispersion should be "smooth", and if that is what Nuyes meant by "precise directional control" then I would agree, but I feel it should be noted that this speaker's dispersion width runs on the narrow side and that is not necessarily a positive. Frankly, I would not be buying this speaker because narrow sweet spots are such a pita.

Neumann are narrow too..

 

[Vacceo]

My point is that there is no indication in ALL the other reviews on this site that narrower dispersion is better when you look at most of the highest scoring speakers. None of those speakers have dispersion this narrow, and in fact Amir often remarks that it is a positive that the dispersion is at least 50 degrees. And yet, here several people are suggesting that this speaker is superlative because it has narrow dispersion. I agree that it is a matter of rooms and preferences, but I do not agree that this speaker is broadly "better" than other speakers and in my experience speakers like this are difficult to position and have narrow sweet spots from which you can enjoy the stereo illusion, making them unsuitable for many listening rooms and environments. I think we all would agree that the dispersion should be "smooth", and if that is what Nuyes meant by "precise directional control" then I would agree, but I feel it should be noted that this speaker's dispersion width runs on the narrow side and that is not necessarily a positive. Frankly, I would not be buying this speaker because narrow sweet spots are such a pita.

I personally like dispersion around 50 degrees tops. That comes from the fact that I use a multichannel system and it is easier to integrate due to better predictability of reflections. Narrow speakers are also a bit more forgiving on placement.

Of course all these traits may be an absolute con for you and the room you´ll be using the speakers. What we both can agree is that a good directivity is any that keeps the response as homogenous as possible between the intended angles.

 

[ROOSKIE]

In terms of the R3 Meta it sure seems like we can't really know about the dispersion character until the Klipple measures it. I'd be surprised if it is Narrower than the original R3 and that speaker is well loved despite the fairly narrow dispersion.

For myself, so far I have expereinced great 'Imaging, Soundstage and Spaciousness' with all types of dispersion widths.

To overgeneralize, in my 2 channel stereo system(not multi channel)=

For music with a lot of recoreded spaceousness I find Narrow Dispersion works very well at reproducing what the recorded space was. Like a direct link into the recording, more like headphones.
Narrow Dispersion does not add much of a sense of spaciousness to recordings that rely on the speaker and playback senario to provide spaciousness.
Imaging is sometimes but not always a bit more pinpoint though sometimes this can feel fake and not lifelike and others times just wonderfull.
The Klipsch RP-600m did not have pinpoint imaging like the KEF R3 yet both have a Narrow dispersion. They sounded very different even after PEQ.

Wide dispersion adds a sense of spaciousness and physical dimension to otherwise less spacious recordings and recordings where there is no real space in the content.
This is great in a lot of cases and yet sometimes can make a recording sound a bit 'euphonic' or overly spread out.
Imaging is often very 'life sized' and while not pinpoint it makes for great scale and ambiance. I tend to prefer imaging that is more spacious and dimensional large vs tightly controlled and 'pinpoint'.
The BMR is a great wide speaker that I have.

Average or Medium dispersion seems to be a good compromise with the character of both of the above. This is made even better for me when that medium dispersion has good spread up into the very high frequencies. Such as with a waveguide like the JBL 4309 uses. The 4309 is just above narrow at 50/55degrees but it holds this very evenly and up to a very high frequency so in a real sense it is not narrow as it is wider than many speakers when averaging the whole spectrum. I really like this.

The dude from Audioholics seems to have had very similar expereinces to mine if you want a more robust source, quoting the site ---

Understanding Loudspeaker Review Measurements Part I

Audioholics’ speaker reviews often contain detailed measurement graphs. In this article, we explain frequency response and the set of curves known as the 'Spin-O-Rama' to help you pick better product.

www.audioholics.com

"As for the qualitative differences between a highly directional speaker and a more non-directional speaker, research done by Canada’s National Research Council (aka. NRC) hints that most people seem to prefer the sound of wide dispersion loudspeakers because the sidewall reflections add a sense of spaciousness and a wider soundstage.1 There was some evidence to suggest that wide dispersion loudspeakers do sacrifice some precision in stereo imaging in order to achieve their greater sense of ambiance, though conclusive research is not in at this time. Ultimately, this would appear to be a matter of personal preference; those who want a greater sense of spaciousness should look for speakers with a low directivity index, and those who want more precise imaging should look for a high directivity index. It should be noted at this point that room acoustics and speaker placement will have a major role to play in these qualities as well. In my own experience, I have found that both wide dispersion speakers and narrow dispersion speakers can have excellent imaging as well as a sense of spaciousness and ambiance."

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

Personal preferences and listening style/way your brain works will affect all of this of course but food for thought...

It is not unusual for Narrow Dispersion speakers to be used in very Large rooms.

In a small room a Narrow Dispersion will put the listener in a position to receive 1st reflections that are from way off axis. There will also be reinforcement in the bass, midbass and lower frequencies that is significantly elevated compared to the mids and highs. This may not work well for some listeners.

Wide Dispersion in a smaller room might keep the off axis 1st reflection closer in tone and power to the original direct sound. This might be better for tonality, while perhaps weaking the sense of exacting imaging.

Wide Dispersion in a large room may not have much effect. If the listener is in a large farfield from the speakers the actual heard Early Reflections start to come from narrower and narrower angles. Thus eventually the effects of wide dispersion could potentially lose their value.

Who knows?? Listening I guess.

 

[mj30250]

Who knows?? Listening I guess.

Pretty much. There is no "best". It's "best for the listening environment and preferences of the listener(s)".

I have Neumann KH 150s on my office desk. When seated at the desk, their well controlled but narrow directivity is very helpful in mitigating issues created by the nearby walls and desktop surface. Imaging is sharp and their projected scale is sufficiently large while never becoming overwhelmed by an excess of reflected sound. The speakers absolutely excel in this application.

However, if I walk straight back and into an adjoining room, while the speakers still sound fantastic and can go loud enough to get even the next door neighbors dancing, the soundstage progressively collapses until it seems as if the music is being generated by a magic bubble floating above the desk rather than live performers. It's a fun little effect but it's not what I'm after in mid / far-field use. I would not be interested in replacing the much wider-dispersion speakers used in my main setup with the Neumanns, even if the Neumanns might otherwise sound just as good.

 

[ROOSKIE]

Pretty much. There is no "best". It's "best for the listening environment and preferences of the listener(s)".

I have Neumann KH 150s on my office desk. When seated at the desk, their well controlled but narrow directivity is very helpful in mitigating issues created by the nearby walls and desktop surface. Imaging is sharp and their projected scale is sufficiently large while never becoming overwhelmed by an excess of reflected sound. The speakers absolutely excel in this application.

However, if I walk straight back and into an adjoining room, while the speakers still sound fantastic and can go loud enough to get even the next door neighbors dancing, the soundstage progressively collapses until it seems as if the music is being generated by a magic bubble floating above the desk rather than live performers. It's a fun little effect but it's not what I'm after in mid / far-field use. I would not be interested in replacing the much wider-dispersion speakers used in my main setup with the Neumanns, even if the Neumanns might otherwise sound just as good.

Interesting.
I'd suggest that in a larger space the speakers would be farther apart so that dispersion width could still work.(or even in the current space but speading them out)
The KH 150 is below average in dispersion pathern width but not that narrow.
Maybe the closeness on the desk helpd foster that soundstage? IDK.

As mentioned in my previous post as the space gets larger the energy concetraion effectively becomes wider.
I.E. standing 3 feet away from the speaker 40degrees of beam width spread is 2feet wide
(50d is 2.8ft wide @3ft from spkr)
(60d is 3.5ft wide @3ft from spkr)
(80d is 5ft wide @3ft from spkr)
standing 11 feet away now the 40degree beam with is 8feet wide per speaker.
(50d is 10feet wide @11ft from spkr)
(60d is 12feet wide @11ft from spkr)
(80d is 18feet wide @11ft from spkr)

30d is only 6feet at 11feet from the speaker. Yikes.

So in a large space a 40d beam width could become very full, but would surely loose some reflection provided spaciousness compared with 60d and 80d beam widths.

 

[cavedriver]

Neumann are narrow too..

I would be inclined to agree with you if it were true but the KH420 review comments on the dispersion being "a bit narrow" while also saying, "+/- 55 degrees" at -6 dB. The graphic suggests it drops down almost to +/- 40 degrees but only above 10k. This was actually one of the reviews I looked at when I suggested that Amir's language suggests that he feels that less than 50 degrees is a negative:

 

[mj30250]

Interesting.
I'd suggest that in a larger space the speakers would be farther apart so that dispersion width could still work.(or even in the current space but speading them out)
The KH 150 is below average in dispersion pathern width but not that narrow.
Maybe the closeness on the desk helpd foster that soundstage? IDK.

As mentioned in my previous post as the space gets larger the energy concetraion effectively becomes wider.
I.E. standing 3 feet away from the speaker 40degrees of beam width spread is 2feet wide
(50d is 2.8ft wide @3ft from spkr)
(60d is 3.5ft wide @3ft from spkr)
(80d is 5ft wide @3ft from spkr)
standing 11 feet away now the 40degree beam with is 8feet wide per speaker.
(50d is 10feet wide @11ft from spkr)
(60d is 12feet wide @11ft from spkr)
(80d is 18feet wide @11ft from spkr)

30d is only 6feet at 11feet from the speaker. Yikes.

So in a large space a 40d beam width could become very full, but would surely loose some reflection provided spaciousness compared with 60d and 80d beam widths.

If you're after that big / spacious soundstage, what's the ideal ratio of dispersion to listening distance to sidewall distance, etc etc? And how do we weight the ratio to account for a speaker with reflected sound that closely matches the direct sound in tonality versus one that makes a mess of reflections?

And while I fully agree that if I was looking to widen the sound stage of the Neumann's while listening to them in the neighboring room, positioning them further apart would help, but I doubt that this would fully close the gap versus my wider-dispersion speakers because the "good" reflections would remain less prominent, as you say.

So yes, I think it basically goes back to looking at your room and your preferences, and then listening.

 

[pablolie]

Am I the only one that would like to see the estimated in-room response, like in other speaker tests in here?

 

[ROOSKIE]

Am I the only one that would like to see the estimated in-room response, like in other speaker tests in here?

He is not using a Klipple.
These are gated and taken in a small room.

I am curious, why do you want to see the estimated in room? That is kind of a pointless one isn't it?
Well, if you like it, I suppose he could do the calculations for predicted in-room if he has all of the necessary angles to calculate the 1st reflections.
He could give an actual room example as well. His room.
I don't know though, I am sticking with the idea that the in-room steady state predicted response is pretty pointless.

 

[mj30250]

He is not using a Klipple.
These are gated and taken in a small room.

He is using a Klippel, but their distortion analyzer, not the full NFS.

 

[pablolie]

....
He could give an actual room example as well. His room.
I don't know though, I am sticking with the idea that the in-room steady state predicted response is pretty pointless.

I kinda disagree - I think it allows one to really establish a comparison in the real world. The measurements that dissect woofer, tweeter etc response may be of interest to a speaker designer, but -personally and selfishly- as a user I'd rather see in-room response that captures the cumulative result... and that was very common in the measurement charts I have seen for speakers in this site before.

And yes - even if it is a particular room, as long as it is the same room for different speakers, it is a good overall reference to have.

It was the one measurement that allowed me to quickly plot/superimpose two different speakers and let me establish for myself whether I think I'd hear a difference, as in the example below. I know it is an intellectual shortcut, but imo a quick and dirty way to establish whether one should explore further...

 

[Descartes]

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

Plus when you have nine speakers at ear level and six in the ceiling having very wide dispersion ends up with cacophony!

 

[Newman]

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

“I don't recall ever saying explicitly that narrow directivity has ANY special virtue (except, obviously, in sound reinforcement and cinema systems). …in many years of subjective testing, wider dispersion and the consequent lateral reflections have been approved of in stereo listening, mostly as a means of softening the image of hard panned left and right sounds (a section of an orchestra emerging from a single loudspeaker is unrealistic).” - Floyd Toole

A secondary reason for wider over narrower dispersion that Dr Toole has mentioned is to soften the Phantom Centre HRTF Issue in 2-channel playback.

As you can see, both these reasons are related to 2-channel playback problems, and it is easy to see that multichannel playback is less demanding on dispersion width, while still rewarding spectral consistency between off-axis and on-axis FR.

OTOH Dr Toole has also mentioned that extreme dispersion eg omni has consequences in loss of imaging localization that make them less preferred, so there is a limit to the ‘wider is better’ idea. But you won’t find it with typical front-firing designs.

cheers

 

[ROOSKIE]

I kinda disagree - I think it allows one to really establish a comparison in the real world. The measurements that dissect woofer, tweeter etc response may be of interest to a speaker designer, but -personally and selfishly- as a user I'd rather see in-room response that captures the cumulative result... and that was very common in the measurement charts I have seen for speakers in this site before.

And yes - even if it is a particular room, as long as it is the same room for different speakers, it is a good overall reference to have.

It was the one measurement that allowed me to quickly plot/superimpose two different speakers and let me establish for myself whether I think I'd hear a difference, as in the example below. I know it is an intellectual shortcut, but imo a quick and dirty way to establish whether one should explore further...

Sure, I completely understand the draw to simplify.
All of that would be true IF the in room response was the 'cumulative result' of what you hear. It is not that, not at all.

Unfortunately, that is the problem with it, at least as I see it. Folks often want it to be that and go and interpret it as that, however the in-room steady state response is not what you hear and it is especially not what you hear with music which is much more dynamic vs a test signal that is steadily emitting the same power.

WIth a test signal it is just the equilibrium of the SPL by frequency in the room at the listening position received by a microphone. (and with the predicted version it is calculated from the early reflections and listening window and no late reflections) This is absolutely not the accumulation of what you hear.
It does relate to the perception of the overall tonality but even in that regard is only a part of that experience.

Basically one must also look at the data in more granular fashion, speaker designer or not. You can not tell what a speaker sounds like from the in room estimated steady state response.

The Spinorama IS the simplified data. That is why it was created to simply vast amounts of data. Same with the various Polars and radars. Those charts are as simple as it can get and still be informative. The fact it can somewhat accurately predict the in-room steady state is a fun side effect, but the reverse is not true. The in-room can not predict the spinorama and spinorama is essentially how the speaker sounds - the in room can not predict how the speaker sounds.

I am not saying don't enjoy the predicted in room graph if you do, nor that it is meaningless, it isn't part of the spinorama for a reason though and may not correlate well with what the speaker actually sounds like (speakers with very similar 'in-room' steady state measurements could actually sound very different and speakers with somewhat notably different 'in-room' steady state SPL levels could in some cases actually sound quite similar)

 

[Newman]

Sure, I completely understand the draw to simplify.

He is not simplifying if he wants an additional graph to those already presented.

 

[pablolie]

He is not simplifying if he wants an additional graph to those already presented.

In all fairness, one that used to be very standard in previous measurement protocols. I also hinted at the graphs that -to me, at least- are pretty meaningless, and don't dispute the possibility my preferences may be mistaken.

 

[mglobe]

I kinda disagree - I think it allows one to really establish a comparison in the real world. The measurements that dissect woofer, tweeter etc response may be of interest to a speaker designer, but -personally and selfishly- as a user I'd rather see in-room response that captures the cumulative result... and that was very common in the measurement charts I have seen for speakers in this site before.

And yes - even if it is a particular room, as long as it is the same room for different speakers, it is a good overall reference to have.

It was the one measurement that allowed me to quickly plot/superimpose two different speakers and let me establish for myself whether I think I'd hear a difference, as in the example below. I know it is an intellectual shortcut, but imo a quick and dirty way to establish whether one should explore further...
View attachment 267201

100% agree.

 

[BrokenEnglishGuy]

“I don't recall ever saying explicitly that narrow directivity has ANY special virtue (except, obviously, in sound reinforcement and cinema systems). …in many years of subjective testing, wider dispersion and the consequent lateral reflections have been approved of in stereo listening, mostly as a means of softening the image of hard panned left and right sounds (a section of an orchestra emerging from a single loudspeaker is unrealistic).” - Floyd Toole

A secondary reason for wider over narrower dispersion that Dr Toole has mentioned is to soften the Phantom Centre HRTF Issue in 2-channel playback.

As you can see, both these reasons are related to 2-channel playback problems, and it is easy to see that multichannel playback is less demanding on dispersion width, while still rewarding spectral consistency between off-axis and on-axis FR.

OTOH Dr Toole has also mentioned that extreme dispersion eg omni has consequences in loss of imaging localization that make them less preferred, so there is a limit to the ‘wider is better’ idea. But you won’t find it with typical front-firing designs.

cheers

Isn't a virtue for itself, the virtue is when you do the Mandatory acoustic treatment for get a good SQ, in narrow dispersion you don't need to put acoustic treatment in the sidewalls, while if you put a wide dispersion speaker in the small room you are going to need a lot of acoustic treatment.







It's easier to work with, that's the '' virtue ''. I don't think the listening room from Dr toole was his bedroom with his PC in there, for example
The wide dispersion are going to '' reflect '' more in the room.


The context for that answer was very different to the case in discusion.

I believe Toole recommends narrow directivity speakers for multichannel. He recommends wide directivity for two-channel, because his starting position is that two-channel should be turned into faux multichannel by bouncing reflections around the room, in search of so-called envelopment.

 

[mglobe]

Sure, I completely understand the draw to simplify.
All of that would be true IF the in room response was the 'cumulative result' of what you hear. It is not that, not at all.

Unfortunately, that is the problem with it, at least as I see it. Folks often want it to be that and go and interpret it as that, however the in-room steady state response is not what you hear and it is especially not what you hear with music which is much more dynamic vs a test signal that is steadily emitting the same power.

WIth a test signal it is just the equilibrium of the SPL by frequency in the room at the listening position received by a microphone. (and with the predicted version it is calculated from the early reflections and listening window and no late reflections) This is absolutely not the accumulation of what you hear.
It does relate to the perception of the overall tonality but even in that regard is only a part of that experience.

Basically one must also look at the data in more granular fashion, speaker designer or not. You can not tell what a speaker sounds like from the in room estimated steady state response.

The Spinorama IS the simplified data. That is why it was created to simply vast amounts of data. Same with the various Polars and radars. Those charts are as simple as it can get and still be informative. The fact it can somewhat accurately predict the in-room steady state is a fun side effect, but the reverse is not true. The in-room can not predict the spinorama and spinorama is essentially how the speaker sounds - the in room can not predict how the speaker sounds.

I am not saying don't enjoy the predicted in room graph if you do, nor that it is meaningless, it isn't part of the spinorama for a reason though and may not correlate well with what the speaker actually sounds like (speakers with very similar 'in-room' steady state measurements could actually sound very different and speakers with somewhat notably different 'in-room' steady state SPL levels could in some cases actually sound quite similar)

I guess I’m misinformed, but I thought the in-room was in imprecise terms, the product of the spinorama data. Essentially the FR at the LP in a theoretical room. Not your room of course, but a consistent theoretical room from test to test. I look forward to being corrected on this.