Controlled vs wide dispersion in a normal living room environment..

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
After years of being frustrated with my loudspeaker/ living room interface, I have come to the conclusion that wide(or even moderately wide) dispersion loudspeakers are not a good thing in untreated "normal" living rooms. When I go to audio shows, I notice that the wide dispersion speakers almost always sound bad if the room is left untreated. The latest example was the otherwise amazing Revel salon 2 at the Newport show. I couldn't wait to get out of the room. All the wide dispersion speakers sounded that way unless they had sidewall treatment. On the other hand, a speaker like the Revel Salon 2 (Or a Radialstrahler, for that matter) in a correctly treated, dedicated, living room is almost unassailable. If you can't treat the room(As in wife?), do whatever else you can to bypass it( i.e., by controlling dispersion)..(By the way,this post is meant mainly for newbies)
 
Well, the problem with most speakers is that they are omni at low and mid-bass, then directional at mids and highs. This gives an unnatural presentation.
So, I think either they have to be completely directional, or completely omni.
Which is hard to do for either case.
I'm in the linkwitz camp myself. I believe in even power response.
If the FR is split into directional and omni, we will need baffle step correction and that messes up the power response. Then the reflections won't be copies of the original, only copies of the lower half of the spectrum.
If we keep an omni far enough from the boundaries, and sit closer to the nearfield, the reflections shouldn't muddy the original, only add spaciousness.
 
Yes, bass is omnidirectional (except in a dipole figure of eight), But if you can adjust the bass level to equal the power response of the controlled high frequencies, you can make a bad situation(waf) a lot better. Sitting in the near field is what I had to do for many years. I really got tired of that limitation..
I think a power response that evenly and gradually slopes down, sounds best with a flat(On axis) measuring speaker, and a flat, even power response sounds best with a speaker who's on axis response gradually slopes down(The one that Linkwitz favors. 2db per decade slope.) This is in a normal living room situation of course. Both a flat on axis response and a flat power response is just way to much high frequency energy for a normal(untreated) living room to handle.
 
Last edited:
Yeah, dipoles are great. They avoid sidewall reflections while maintaining even power response.
I don't think it's possible to adjust bass and midbass to maintain even power response without compromising the direct FR.
I like a downward slope as well because it adds distance to the image.
As for the nearfield, I think it gets fatiguing after a while because I need to hear the room for a natural soundfield.
The reason for this is that I am accustomed to the sound of my room, so when a system presents a soundfield that differs from what I am accustomed too, it sounds foreign to me.
 
Well, the problem with most speakers is that they are omni at low and mid-bass, then directional at mids and highs. This gives an unnatural presentation.
So, I think either they have to be completely directional, or completely omni.
Which is hard to do for either case.
I'm in the linkwitz camp myself. I believe in even power response.
If the FR is split into directional and omni, we will need baffle step correction and that messes up the power response. Then the reflections won't be copies of the original, only copies of the lower half of the spectrum.
If we keep an omni far enough from the boundaries, and sit closer to the nearfield, the reflections shouldn't muddy the original, only add spaciousness.

Two things: recordings are commonly mixed with speakers that are "omni at low and mid-bass, then directional at mids and highs". What happens when you play back such a recording over constant directivity speakers?
Secondly how far is "far enough from the boundaries"?
 
another ouroboros thread. How many of us are really able to listen in the far field anyway? For many particularly in the UK, listening room size dictates the listening field. With OB its even worse a scenario. So as interesting as the subject is, its simply out of the realms of possibility for quite a few folks. This would explain the classical use of wide pattern radiation and room treatment, which seems to fit better to the modest living spaces we have in the UK. (unless your a multimillionaire)
 
"Farfield" is a relative term too, I guess. Here's my limited dispersion setup. Neo 10 dsp'd flat at 60", on axis(Of course). All monopole..
 

Attachments

  • 009.jpg
    009.jpg
    383.6 KB · Views: 1,154
Last edited:
The dispersion doesn't actually matter too much on the amount of direct to reflected sound you hear in a listening room. Reflected sound will always dominate being between 3dB and 6dB higher for pure monopole speakers. With a typical value of 5dB dominance of reflected sound for wide dispersion monopole speakers.

Most people will in blind prefer speakers turned away 180 degrees when listening to strictly instrumental or effect driven music or material, and prefer speakers turned "correctly" when listening to speech or vocal (non-chorus) dominated music or material. With a normal mix of popular music the preference is in general random on blind tests.
 
Are you Amar Bose in disguise?

No. I must decline that honor. My work in the area has also focused on the refined work of Dr. Floyd Toole of Harman Research.

But it's a simple experiment to make at home. The emphasis is that it must be a pure blind test. Those participating must not know your speakers or listening room beforehand. And they must not know, or have any indication when a speaker is turned around or not.

Go ahead and try. You will find the same result as everyone has.
 
Last edited:
I had the pleasure of participating in this kind of test once. Facing rear sounded like garbage with poor localization and terrible unevenness in the top end. It might depend on your hearing. I still have good HF hearing. It made an otherwise reasonable speaker sound like, well a Bose. Muffled, fuzzy and other wise what 9 49-cent drivers would be expected to sound like no matter how much eq was forced on them.

The parameter that effects the dispersion the most in my experience is edge diffraction. I have heard many MTM speakers, but I am not sure they were actually built as the good Dr. Joe said. Might have to build some to see.
 
Two things: recordings are commonly mixed with speakers that are "omni at low and mid-bass, then directional at mids and highs". What happens when you play back such a recording over constant directivity speakers?
Secondly how far is "far enough from the boundaries"?

Well, I think mixing rooms are set up with monitors operating in 2¶ space, and I think they do that for this reason. Could you imagine listening in a car or on headphones if they added BSC?
As far as room boundaries, I don't subscribe to a set distance (haas, precedence effect) I think its a matter of ratios. The closer I am to the source, the closer that source can be to boundaries. So I guess, close enough for distinct echo (beyond precedence) but far enough where the room can still contribute.
 
controlled pattern obviously offers benefits, it just depends what the 'pattern' is, for the particular room. In my L shaped listening room of 4m x 4m x 3m, my listening space is an equilateral triangle of about 60", so im almost never close to farfield distances. Options are pretty much limited to monopole only. Keysers has a thread: 'listening near the back wall' or something to that effect. That thread fits my scenario perfectly. In my small room the difference in switching from dome to ribbon was profound, the lack of ceiling reflection was a huge improvement. In this case the drivers were Visaton g20sc 20mm fabric dome, and fountek neoCD3.5H horn loaded 'ribbon'. Both are around 30GBP and substantially flat on axis in their passband. Both sound great, the ribbon just present the room in a better way.
 
The dispersion doesn't actually matter too much on the amount of direct to reflected sound you hear in a listening room. Reflected sound will always dominate being between 3dB and 6dB higher for pure monopole speakers. With a typical value of 5dB dominance of reflected sound for wide dispersion monopole speakers.
I have to disagree with this.

An informal test that was conducted in another thread a few months ago (I don't recall which thread off hand without searching for it or I would post the link. Markus might remember which thread) using a reverse phase cancellation method of measuring the reverberant field found that of the participants the direct field was between 3dB and 10dB stronger than the reverberant field through the midrange and treble at the listening position, not the other way around. All or nearly all systems tested were monopoles from memory.

Of course these were fairly well set up systems compared to usual consumer fare dumped into the corners of a room, and the ones getting 10dB were using horns/waveguides...but it served to illustrate the point that when an experienced listener perceived that they were listening within the near field that the direct signal was indeed at least 3dB or so stronger than the reverberant field.
 
Last edited:
I had the pleasure of participating in this kind of test once. Facing rear sounded like garbage with poor localization and terrible unevenness in the top end. It might depend on your hearing. I still have good HF hearing. It made an otherwise reasonable speaker sound like, well a Bose.

It's actually irrelevant what your experience is as there would be as more subjects with the reverse perception. What is tested is the general perception of sound, not that of the individual subject.

A more comprehensive experiment was made by construction a decent quality system. And use a bipolar speaker where the speaker was effectively split down the middle. In that way, using a balance control you could ensure equal total output but varying degrees of reverse sound could be generated. In this experiment, none of the test subjects preferred pure monopole sound, all preferred some but varying degree of reverse sound depending on material.
 
I'm with tvrgeek on this one. A reverse facing monopole pointing into the corner of the room sounds like garbage. A perfect example is flat screen TV's with rear facing speakers. The speakers used in such TV's are bad enough when facing forwards but become almost unusable when facing backwards with atrocious dialogue clarity, let alone quality on music.

Why does it sound bad ? I would say part of the reason is that the image producing high frequencies are delayed behind the low frequencies. The directional high frequencies can only reach us by bouncing off the wall behind and travelling a significantly further distance, while the low frequencies (middle midrange and down depending on baffle size...) can simply wrap around the cabinet and arrive directly - well before the high frequencies.

Thus massive group delay at high frequencies. Regardless of what the "general perception of sound" of the unwashed masses may be, how can this huge amount of group delay be a good thing ? Then there's the cavity resonance between the back of the speaker and the corner behind it to name one of many other problems...

I would also point out that the masses prefer over bright over blue pictures on their TV's, and audio with boosted bass and treble, does that make it correct or better quality ?
 
Last edited:
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.