Why Lower Crossover Frequency Does Not Always Improve Loudspeaker Directivity

[Jonazkur]

Strange, I don´t see that. There is a -2.5dB dip around 2.7K and a -2dB @4K on axis, but both very narrow-banded.



I see no evidence of diffraction here except for the little 4K ripple which might correspond with the intentional edge diffraction at the sub baffle´s edges.

Trying to judge directivity from just three different angles is a bit difficult, but all the peaks and dips you are mentioned, are pretty narrow-banded and alternating, which hints to the midrange open baffle cardiod as a cause. Not sure if this 1.5K behavior is intentional, but I wouldn't really worry about it. After all, perceived tonality of reverb is not a matter of straight lines on a graph, but rather of the relative SPL of neighboring octave-broad bands (or broader). Particularly around 1K and below it looks pretty good, and that is where usually problems originate from.

well to me the off axis looks really strange as there is 7-8db deviations. You can pretty well see the behavior with those angles. Polar heatmap would be easier to judge.

 

[Waveform Fidelity]

Well, the LS50 is modified to play a little bass, whereas the R3 is a dedicated midrange.. soo
I definitely use a small dedicated chamber for my R3 coax.... Almost copy paste from the R900... Just better coax, lower cross over and bigger woofers.

Interesting - whats the raw frequency response with your R3 Meta and R900 volume? Any peaking in the LF roll-off which I assume you see as a nice -12dB/Oct acoustic roll-off

 

[Arindal]

the off axis looks really strange as there is 7-8db deviations.

No doubt about the existence of this off-axis phenomenon, but this is not affecting the listening window, nor are there any hints it is caused by diffraction. I would rather suspect an effect of the midrange cardioid, might be intentional or unintentional, might show the opposite behavior vertically (because of the sub-baffle). The manufacturer claimed to intentionally implement HRTF-related curves on the sound power curve in the room, so it does not really surprise me (although I do not understand the purpose in this case).

Despite from looking ugly on the 60 and 90deg horizontal graph, what exact degradation of sound quality would you expect here? Have heard these models solely under top-notch studio control conditions, and they sounded as neutral as can get.

 

[Jonazkur]

No doubt about the existence of this off-axis phenomenon, but this is not affecting the listening window, nor are there any hints it is caused by diffraction. I would rather suspect an effect of the midrange cardioid, might be intentional or unintentional, might show the opposite behavior vertically (because of the sub-baffle). The manufacturer claimed to intentionally implement HRTF-related curves on the sound power curve in the room, so it does not really surprise me (although I do not understand the purpose in this case).

Despite from looking ugly on the 60 and 90deg horizontal graph, what exact degradation of sound quality would you expect here? Have heard these models solely under top-notch studio control conditions, and they sounded as neutral as can get.

Diffraction can happen off axis too. If you just listen these in a dead dampened room with no soundstage then prob its not huge problem but for normal listening it looks bad. I have no way telling how this would sound in my room but surely its not neutral as those 40 - 80 degree reflections are also important. If you have good listening window and then huge problems after that it hints some problems like diffraction or odd driver behavior.

 

[Arindal]

Diffraction can happen off axis too.

Surely, but diffraction usually leaves identifiable traces like countering cancellation/addition dips/peaks, if you compare several response graphs taken at different angles. Zero signs of such here below 2K, so I suspect the midrange cardioid.

If you just listen these in a dead dampened room with no soundstage then prob its not huge problem but for normal listening it looks bad.

I agree it looks bad on the graph, but it did not sound anyhow colorated or bad in a studio environment, which was more reverberant than most of control rooms I know. Maybe anyone has listened to them under home conditions and can report.

I have no way telling how this would sound in my room but surely its not neutral as those 40 - 80 degree reflections are also important.

99.9% of all speakers on the market are not neutral under 40-80deg, most of them rather heavily and broad-brandedly colorated (particularly those being frequently tauted and recommended on this board). So why would a rather narrow-banded boost in off-axis energy in a rather uncritical frequency band cause such criticism without anyone being able to explain which exact sound quality degradation it is causing?

 

[Jonazkur]

Surely, but diffraction usually leaves identifiable traces like countering cancellation/addition dips/peaks, if you compare several response graphs taken at different angles. Zero signs of such here below 2K, so I suspect the midrange cardioid.



I agree it looks bad on the graph, but it did not sound anyhow colorated or bad in a studio environment, which was more reverberant than most of control rooms I know. Maybe anyone has listened to them under home conditions and can report.



99.9% of all speakers on the market are not neutral under 40-80deg, most of them rather heavily and broad-brandedly colorated (particularly those being frequently tauted and recommended on this board). So why would a rather narrow-banded boost in off-axis energy in a rather uncritical frequency band cause such criticism without anyone being able to explain which exact sound quality degradation it is causing?

Most of the good speakers i have heard are smooth to 90deg. Point is IMO this speaker looks bad with its dispersion and honestly looks really strange with its behaviour. But cant say 100% whats causing it without getting it to myself and measuring. Sure it can sound good to some still.

 

[Arindal]

Most of the good speakers i have heard are smooth to 90deg.

How is ´smooth´ defined in acoustic terms, and what makes you certain that ´non-smooth´ inevitably sounds bad? I agree the horizontal FR look strange, but why is this bad in terms of sound? Or are we solely discussing geometric beauty of the graphs?

I am highlighting this in a thread on directivity, because a lot of people seem to believe that ´smooth directivity´ equals continuously increasing directivity index, therefore being satisfied with some more or less straight, but tilted-downwards lines on the graph. I have never in my life heard a good speaker with such behavior except under really overdamped conditions, close to irrelevancy of indirect sound. If you look at the heavy, broadbanded colorations of reflections and reverb in this case (usually very heavy on lower midrange and midrange, in worst cases severely lacking energy between 2.5 and 8K in the room), it becomes apparent what sounds wrong here, and I can predict it easily looking at the graphs (despite from being cautious with predictions which are solely based on measurements).

The majority of speakers I have heard which sounded excellent and created a plausible reverb imaging/ambience, were more or less resembling constant directivity speakers over octave-broad or broader bands, particularly showing similar midrange directivity behavior (300-1.5K) compared to what they did between 2.5K and 8K, in terms of overall directivity index, side wall reflections, floor, ceiling and console/desk bounce. I don´t recall an example of narrow-banded directivity issues which really sounded bad, so I tend to ignore them, rather looking at broad-banded behavior.

 

[markstein]

Despite from looking ugly on the 60 and 90deg horizontal graph, what exact degradation of sound quality would you expect here? Have heard these models solely under top-notch studio control conditions, and they sounded as neutral as can get.

The question is: how do these loudspeakers sound in a typical listening room with reflective walls (not top-notch studio conditions)? In a well damped room, loudspeakers with suboptimal directivity will of course sound much better as the negative effects will be addressed by acoustic room treatment.

 

[Jonazkur]

How is ´smooth´ defined in acoustic terms, and what makes you certain that ´non-smooth´ inevitably sounds bad? I agree the horizontal FR look strange, but why is this bad in terms of sound? Or are we solely discussing geometric beauty of the graphs?

I am highlighting this in a thread on directivity, because a lot of people seem to believe that ´smooth directivity´ equals continuously increasing directivity index, therefore being satisfied with some more or less straight, but tilted-downwards lines on the graph. I have never in my life heard a good speaker with such behavior except under really overdamped conditions, close to irrelevancy of indirect sound. If you look at the heavy, broadbanded colorations of reflections and reverb in this case (usually very heavy on lower midrange and midrange, in worst cases severely lacking energy between 2.5 and 8K in the room), it becomes apparent what sounds wrong here, and I can predict it easily looking at the graphs (despite from being cautious with predictions which are solely based on measurements).

The majority of speakers I have heard which sounded excellent and created a plausible reverb imaging/ambience, were more or less resembling constant directivity speakers over octave-broad or broader bands, particularly showing similar midrange directivity behavior (300-1.5K) compared to what they did between 2.5K and 8K, in terms of overall directivity index, side wall reflections, floor, ceiling and console/desk bounce. I don´t recall an example of narrow-banded directivity issues which really sounded bad, so I tend to ignore them, rather looking at broad-banded behavior.

Smooth is just without ripples doesnt matter if its constant or not it just shows good low diffraction behaviour and good directivity matching. Listening distance and rooms are different so different DI profiles should exist ofc. If you have 120 degree constant radiation it will sound prob bright in normal reflective room. And i agree that broadband DI is more important than narrow variations but its no excuse to ignore them. Non smooth usually have sounded worse than smooth ones to me atleast, usually brightness and fatigue in some areas even if on axis is perfectly flat. And the speaker we talk about have big issues imo in 1-3k area and hard to predict the sound.

 

[Arindal]

how do these loudspeakers sound in a typical listening room with reflective walls (not top-notch studio conditions)?

No practical experience with these models under home conditions, only the 12" derivate, which is one of the best speakers I have ever listened to. Don´t see any reason why the big ones would not sound as natural, plausible in imaging and astonishingly realistic in terms of both ambience and depth-of-field, just as they do in studios. Despite from a rather lean and transparent lower midrange (which many recording engineers prefer to judge details at lower SPL, hence the success of Kii Audio), the most obvious thing is exemplary localization of phantom sources in what appears to be a plausible, homogeneous, uncolorated reverb field.

Smooth is just without ripples doesnt matter if its constant or not it just shows good low diffraction behaviour and good directivity matching.

I would agree on the importance of ripples and edge diffraction issues on-axis and within the listening window, but such are not at play here. Off-axis I do not see the point, unless we are talking about very broad-banded colorations. Which ´ripples´ are not, by definition, but close to omnidirectional (lower) midrange paired with narrow treble/brilliance dispersion, is.

Listening distance and rooms are different so different DI profiles should exist

I agree, but different listening distances at given reflexivity in my understanding just call for different yet constant directivity indices.

And yes, rooms are different, but the vast majority of home listening rooms are underdamped in the bass and lower midrange, while some are showing signs of overdamping beyond 5K, so any reasonable DI profile of a speaker should be the exact opposite, should it not? Where are the speakers with very narrow dispersion below 800Hz and decreasing D.I. above 5K? They don´t exist.

If you have 120 degree constant radiation it will sound prob bright in normal reflective room.

Why would it be so? A tonally balanced reverb field sounds, well, as neutral as the one on the recording, and the majority of home listening rooms show decreasing reflexivity/RT60 towards higher frequencies. So no indication of bright-sounding reverb.

Non smooth usually have sounded worse than smooth ones to me atleast usually brightness and fatigue in some areas

Could you name an example of a speaker with is linear on-axis and shows something resembling constant directivity over all localizable bands, with just a narrow band being ´non-smooth´, hence sounding worse, in your experience?

Tbh I do not recall a single one. What usually sounds bad, exactly as you describe it, is free-mounted or non-waveguided tweeters showing a significant decrease in directivity index over a broad band (typically 3-8K, depending on geometry and crossover freq), in comparison with narrow midrange dispersion below crossover frequency (Typically attributed to B&W and similar designs). We can easily agree on that, but it is not at play here.

 

[Jonazkur]

No practical experience with these models under home conditions, only the 12" derivate, which is one of the best speakers I have ever listened to. Don´t see any reason why the big ones would not sound as natural, plausible in imaging and astonishingly realistic in terms of both ambience and depth-of-field, just as they do in studios. Despite from a rather lean and transparent lower midrange (which many recording engineers prefer to judge details at lower SPL, hence the success of Kii Audio), the most obvious thing is exemplary localization of phantom sources in what appears to be a plausible, homogeneous, uncolorated reverb field.



I would agree on the importance of ripples and edge diffraction issues on-axis and within the listening window, but such are not at play here. Off-axis I do not see the point, unless we are talking about very broad-banded colorations. Which ´ripples´ are not, by definition, but close to omnidirectional (lower) midrange paired with narrow treble/brilliance dispersion, is.



I agree, but different listening distances at given reflexivity in my understanding just call for different yet constant directivity indices.

And yes, rooms are different, but the vast majority of home listening rooms are underdamped in the bass and lower midrange, while some are showing signs of overdamping beyond 5K, so any reasonable DI profile of a speaker should be the exact opposite, should it not? Where are the speakers with very narrow dispersion below 800Hz and decreasing D.I. above 5K? They don´t exist.



Why would it be so? A tonally balanced reverb field sounds, well, as neutral as the one on the recording, and the majority of home listening rooms show decreasing reflexivity/RT60 towards higher frequencies. So no indication of bright-sounding reverb.



Could you name an example of a speaker with is linear on-axis and shows something resembling constant directivity over all localizable bands, with just a narrow band being ´non-smooth´, hence sounding worse, in your experience?

Tbh I do not recall a single one. What usually sounds bad, exactly as you describe it, is free-mounted or non-waveguided tweeters showing a significant decrease in directivity index over a broad band (typically 3-8K, depending on geometry and crossover freq), in comparison with narrow midrange dispersion below crossover frequency (Typically attributed to B&W and similar designs). We can easily agree on that, but it is not at play here.

There is problems in listening window area too… big problems and bunching centered to 4khz also 1.5khz from those i would assume it will sound bright and fatiguing. Imo those colorations are broad 1-4khz. Maybe im too critical.

One example of little inconsistency in midrange area is genelec 8361 and i found the sound little bit harsh in that area 2-4khz. Otherwise a good speaker. And ofc in nearfield you dont have that problem.

 

[Arindal]

big problems and bunching centered to 4khz also 1.5khz from those i would assume it will sound bright and fatiguing. Imo those colorations are broad 1-4khz.

Sorry, I don´t see these broad problems and colorations on axis. Not at all.

In studios and broadcast institutions, these monitors are referred to as one of the least fatiguing models of all. Can confirm that from own experience, nothing sounds bright or fatiguing here.

 

[Jonazkur]

Sorry, I don´t see these broad problems and colorations on axis. Not at all.

In studios and broadcast institutions, these monitors are referred to as one of the least fatiguing models of all. Can confirm that from own experience, nothing sounds bright or fatiguing here.

I mean the 10 and 30 degree ones which are in listening window. I clearly see them they bunch up over the on axis line and it will probably cause brightness. If you listen these in studio nearfield then there is no problem.

 

[markstein]

No practical experience with these models under home conditions, only the 12" derivate, which is one of the best speakers I have ever listened to. Don´t see any reason why the big ones would not sound as natural, plausible in imaging and astonishingly realistic in terms of both ambience and depth-of-field, just as they do in studios. Despite from a rather lean and transparent lower midrange (which many recording engineers prefer to judge details at lower SPL, hence the success of Kii Audio), the most obvious thing is exemplary localization of phantom sources in what appears to be a plausible, homogeneous, uncolorated reverb field.

I listened to the ME901K (16" bass) in an unknown "typical" living room and can confirm the lean lower mid character. My Neumann KH310 which were set up as a comparison (no EQ) sounded fuller and voices had more "body" - not sure whether this was due to SBIR or not.
Despite the lean character of the ME901K they did not sound thin and annoying - however, it was difficult for me to come to a final conclusion after my rather short listening session. Would love to have a test pair in my room for two weeks which is not possible without buying.

 

[Arindal]

My Neumann KH310 which were set up as a comparison (no EQ) sounded fuller and voices had more "body" - not sure whether this was due to SBIR or not.

This is to be expected due to different baffle step beginning to narrow down directivity from a certain frequency upwards. A 16" midbass (MEG) is basically offering one full octave more of directional lower midrange dispersion (250-600Hz compared to an 8" midwoofer (Neumann), approximating omnidirectional behavior below 500Hz hence lower midrange heavy reverb in an untreated room. Behavior above lower crossover frequency (around 600Hz in both cases) might play a role here as well, as a 5" open baffle cardioid (MEG) offers significantly higher directivity index compared to a non-waveguided 3" dome (Neumann).