Omni-directional Speakers

[Eric SVL]

I have been curious about omnidirectional speakers for a while now, but have not personally heard any. I wanted to make a thread to discuss some of the designs that have caught my attention and to see if anyone has experienced them.

Ohm Walsh

Linkwitz Lab LXmini

Linkwitz LX521

DECWARE model ERRx

DECWARE model HR-1

GR Research O-3

Duevel Speakers

MBL Speakers

There is also Parallel Audio but I'm unsure if any speakers were ever released.

Note that some of these speakers are more of a hybrid rather than purely omni, as the tweeters are directional, which gives more focused imaging. They are also less picky about placement than pure omni speakers.

I have done some reading, but no listening. What do these kinds of speakers tend to do well? What do they not do well?

 

[Eric SVL]

This was an informative discussion on Audiogon:

https://forum.audiogon.com/discussi...000-series-disrespected-because-of-omni-direc

 

[tesseract]

Yes, my X-Omnis image like crazy, spooky-realism.

 

[Eric SVL]

Yes, my X-Omnis image like crazy, spooky-realism.

I haven't heard of those. Looks like they were designed by Danny Richie? Do you currently use them? I see the foam at your front corners, so probably not.

 

[tesseract]

I haven't heard of those. Looks like they were designed by Danny Richie? Do you currently use them? I see the foam at your front corners, so probably not.

They have not been hooked up in years. They do better in spaces larger than I have available for them at the moment.

 

[Eric SVL]

Normally, when setting up in a new room, I have to spend some time optimizing the midbass. That entails moving the mains and seats and measuring until I've eliminated all the dips and nulls at the seating position. Then I can do the same with the subs and run Audyssey.

I'm intrigued by the idea that due to the design of these speakers, I won't have those dips and nulls to contend with from the mains. Am I understanding this correctly?

 

[tesseract]

Normally, when setting up in a new room, I have to spend some time optimizing the midbass. That entails moving the mains and seats and measuring until I've eliminated all the dips and nulls at the seating position. Then I can do the same with the subs and run Audyssey.

I'm intrigued by the idea that due to the design of these speakers, I won't have those dips and nulls to contend with from the mains. Am I understanding this correctly?

I have never put a mic to these so can't answer to peaks and nulls, but Danny says they need to be at least 3 ft. from any boundary to work correctly. Moving them away from the wall puts them too close to my flatscreen.

I have heard them in an open space and the SS&I is superb. I keep them for the day that I have a dedicated audio room and can set them up as intended.

 

[Matthew J Poes]

Normally, when setting up in a new room, I have to spend some time optimizing the midbass. That entails moving the mains and seats and measuring until I've eliminated all the dips and nulls at the seating position. Then I can do the same with the subs and run Audyssey.

I'm intrigued by the idea that due to the design of these speakers, I won't have those dips and nulls to contend with from the mains. Am I understanding this correctly?

That was the argument that adherents claimed, but the reality is that the peaks and dips, such as from SBIR and room modes will exist regardless. The radiation from these speakers can shift them around a little, but it doesn't eliminate them. Earl Geddes and Linkwitz got into a friendly disagreement about this as it pertained to bass, so Geddes did a study comparing numerous bass radiation patterns and showed that it did not improve the bass response. While it did change how modes were excited, it didn't actually decrease deviance (meaning bass was no flatter or smoother, just different).
At mid and high frequencies the room doesn't contribute all that much to the response shape, it's mostly the direct sound. If anything, a very wide angle speaker with a lot of room reflections might measure worse in the room from the contribution of those reflections to the steady-state measurement.
I think the desirability of omni-polar behavior is really more philosophy. Bill Duddleston, Earl Geddes, Floyd Toole, etc. will all argue that a speaker should be quite the opposite, very tightly controlled dispersion. They will argue this gives the smoothest response, best behaved early reflections, and thus best soundstage, tonal balance, etc. People like Linkwitz, Jim Winey, Amir Bose, Von Schweikert, etc. will all argue that some omni-radiation or bi-polar radiation is beneficial.
I've heard a few speakers on your list, including a bunch of the Decaware speakers. I have my favorites. I happen to agree with the first camp I mentioned and find they offer the most precise soundstage (Constant directivity speakers). I also agree with the science they ascribe too. Having said that, I always try to be open-minded. What drew me to them was bad experiences with speakers that had bi-polar or omni dispersion pattern and what I felt was bad sound. Many had a very diffused and indistinct soundstage. I had a friend who loved the Bose 901's and I always felt those produced this larger than life soundstage that was very unrealistic. It had very vague instrument placement. Magnepans had this head in a vice problem with a similar kind of indistinct soundstage. The Decaware's were just bizarre sounding to be honest. I think those were built for musical fun rather than accuracy. I had a lot of fun hooking those up to some SET amplifiers and finding great tracks to make them sound good, but...putting on a reviewer cap, they weren't a great speaker. Very unnatural tonal balance in my experience with a strange phasie sound to them.
I've heard a few of Linkwitz speakers. They sounded really good.
I think Wayne may have built one of the Linkwitz speakers, so he can comment on those. My experience was at a fellow Audiophiles house, so my time was limited. A few hours maybe at most.

 

[Eric SVL]

Thank you for your comments, Matt.

I think the desire for omnidirectional speakers is more than just philosophy. I read nothing but raves about the Ohm Walsh's, and you said the Linkwitz speakers sounded really good, too. I'm intrigued.

 

[tesseract]

I am a huge MBL fan. Those omnis have no problem throwing up SS&I.

 

[NBPK402]

I wonder how these would work for surround speakers?

 

[Eric SVL]

I wonder how these would work for surround speakers?

I wondered the same, and Ohm does make models specifically for that purpose. I asked them about how they fit in in an industry that is moving away from diffuse surrounds and towards point sources. My impression is that they don't agree with the new specs, arguing that point sources are the opposite of what you want in surround sound. At any rate, their speakers still have a direct-radiating tweeter, so only the walsh driver is omnidirectional.

 

[tesseract]

I wonder how these would work for surround speakers?

Well, if you can get them 3-4 ft. minimum away from boundaries. The X-Omni I posted were originally intended for surround, they just happen to image so well that the 2 channel community adopted them.

 

[Eric SVL]

Well, if you can get them 3-4 ft. minimum away from boundaries.

I'm not sure what alterations are made, if any, but there is a Walsh Wall Mounted Surround Omni:
https://ohmspeaker.com/speakers/walsh-wall-mounted-surround-omni/#wwosat-1000

Pricey for surrounds though.

I added the GR Research O-3 to the list, which is a relative of your X-Omni.

 

[Matthew J Poes]

Toole Loudspeakers and Room Part 1

You might find this article interesting @Eric SVL, it is written by Floyd Toole and talks about all of these topics.

Don't dismiss the article because of his conclusions, it's a good read for what it does say. He does conclude that we don't need omni or dipole radiating speakers in a modern multi-channel era, but he also makes clear why those technologies came about. Given that the vast majority of music is 2 channel and a lot of people aren't interested (or able) to set up a surround system in a manner that is optimal for music, I can totally see why people are still interested in these other arrangements.

He also talks about the merit of dipole surrounds (omni being another approach to "Spraying sound everywhere"), and rightly notes they weren't dipole (or bipole) ever, they were always more of a form of omni radiating speaker. He makes a point that I intuitively felt made sense myself, and had made similar suggestions. For the surround channels themselves, if the room is large, it is better to use many direct radiating speakers. As is used in professional cinemas. Ideally with DSP processing to create additional channels for each additional layer of surrounds by delaying and decorrelating the signal to each successive layer. However, in smaller rooms where you sit closer to the surrounds, these omni-radiating surrounds may be a better option. It is partly why I built the tripole radiating surrounds, which do have fairly even radiation across a very wide angle.

Also, Eric, I just want to note that my comment about "philosophy" wasn't meant to be dismissive. I was simply pointing out that there are many ways to get good sound, and how you choose to go about that is based on your beliefs into what is best. There is not necessarily a right or wrong answer. The idea behind omni and dipole speakers (or any speaker adding ambiance drivers to the sides, rear, or top) is to create a large gap in the initial time delay with a large amount of reflected delayed sound. This gives a perception of spaciousness. It's more consistent with what we would hear in a live event. There are a lot of ways to introduce this. One method is to use a speaker that sprays sound everywhere, as Toole says. This approach naturally caused the increased reflected sound. However if not done well and in a room that is not treated for this, it can create a large amount of reflected energy, but not a large gap in the ITD. This is why I'm not a huge fan of Omnidirectional speakers myself, this is a more common effect with them. Another is to radiate sound in specific directions like behind the speakers or to the ceiling (or both) as this creates a delay in the reflections. Toole himself played around with side radiating ambiance drivers to create this big ITD gap and improve spaciousness. The other way to do this is to recreate the delayed energy by processing the musical signal and sending the delayed reflected sound to ambiance speakers, aka surround sound. In a room with a low RT60, symmetric speaker placement, and good processing, I happen to think this approach is the most natural. The problem typically is that a) the processing isn't always very natural sounding and b)symmetric surrounds in a dead room is highly uncommon.

Please don't let my feelings on this dissuade your thinking though. The great thing about this hobby is that there are so many ways to good sound, and it's fun to experience them all. I'm a scientist by trade and very science minded, so I tend to latch onto ideas that fall into that left brain way of thinking.

 

[Eric SVL]

No offense taken.

The other major benefit is the room-filling sound and large "sweet spot". With these hybrid omni speakers you get the ability to fill an area, not just a single point, with the "best sound" and not be off-axis when you move. So you can go about your cooking, cleaning, working out, seating multiple guests (only one person gets the middle seat!), etc, and nearly always get good sound. It is the opposite of the Magneplanar effect.

 

[Matthew J Poes]

No offense taken.

The other major benefit is the room-filling sound and large "sweet spot". With these hybrid omni speakers you get the ability to fill an area, not just a single point, with the "best sound" and not be off-axis when you move. So you can go about your cooking, cleaning, working out, seating multiple guests (only one person gets the middle seat!), etc, and nearly always get good sound. It is the opposite of the Magneplanar effect.

Sure if that is your goal.

It's worth noting a surround system doesn't have a single "best" sounding seat. The whole point of surround was to expand the optimal number of seats which had a convincing and accurate sound stage. That is why a center channel was added. Stereo speakers have a perfectly good center image, but only for a small area. Setup correctly with the right speakers and many seats will have equally good sound and don't need to be right in the middle. Through the use of constant directivity speakers setup with the correct toe, multiple channels, and appropriate distance, you can take advantage of time-intensity trading to create an enveloping space that is really large. The soundstage remains stable in all of these seating positions. You still need to be inside the acoustic bubble of the speakers, but that bubble can be quite large. In a commercial cinema, for example, 30% of the seats have equally excellent sound. In small spaces like a typical living room we might only get 4-6 equally good seats, but still, that's a lot.

I would really think that even the best omni speakers, which may produce a lot of sound, aren't going to produce an accurate reproduction of the acoustic space everywhere in the room, in fact, I think it would be quite diffused and indistinct in a lot of locations.

 

[Eric SVL]

Sure if that is your goal.

It's worth noting a surround system doesn't have a single "best" sounding seat. The whole point of surround was to expand the optimal number of seats which had a convincing and accurate sound stage. That is why a center channel was added. Stereo speakers have a perfectly good center image, but only for a small area. Setup correctly with the right speakers and many seats will have equally good sound and don't need to be right in the middle. Through the use of constant directivity speakers setup with the correct toe, multiple channels, and appropriate distance, you can take advantage of time-intensity trading to create an enveloping space that is really large. The soundstage remains stable in all of these seating positions. You still need to be inside the acoustic bubble of the speakers, but that bubble can be quite large. In a commercial cinema, for example, 30% of the seats have equally excellent sound. In small spaces like a typical living room we might only get 4-6 equally good seats, but still, that's a lot.

I would really think that even the best omni speakers, which may produce a lot of sound, aren't going to produce an accurate reproduction of the acoustic space everywhere in the room, in fact, I think it would be quite diffused and indistinct in a lot of locations.

I was actually referring to the front stage, not surrounds. Every review I read on speakers like these comments on how the sound doesn't change much when they move around - unlike all the other speakers they are used to. And that includes many higher-end speakers with great sound.

Look at the Ohm Walsh, for example. The tweeters, when aimed properly, project up into a space slightly above and in front of you. This does a couple of things. It creates more life-size imaging (not golfball size), and the sweet spot is expanded laterally. As you move to the right, you get more of the left tweeter. As you move to the left, you get more of the right tweeter. There becomes few "poor" spots as the room is filled with sound and it is harder to find a place that is off-axis. There are even comments that adjacent rooms get more quality sound than they did with their other speakers. That's only good if you want it, but it's just more to add to the point of room-filling sound. And their ability to image is only enhanced by being able to project images within that huge soundstage. I think this is especially important in a small room where you will only sit maybe 9 feet from the speakers, sometimes less. People in the outer seats (of which there may be only 3 or 4 total in the room) are not going to get the same sound experience as those in the center with a purely direct-radiating front stage - they will be off-axis. I think this meshes with your point of direct-radiating speakers doing well in larger spaces where the distances are great enough for things to "come together" better.

The Ohms also use a full-range driver with no crossover, which creates an exceptionally smooth response and integration with the tweeter. There are also remarks about the coherence of sound that the walsh driver creates. So there's more going for it than just the "omni" aspect. It's just really different and needs to be heard.

I don't own speakers like this and haven't heard them myself, but I have read and understand what they do well from other reviewers. I'd be intrigued enough to try a demo if I felt okay about tying up the cash for a few months.

 

[Eric SVL]

I should point out that for my personal setup, I don't run a center channel because my 64" plasma screen doesn't leave room for one. Phantom tends to work best due to the image being created right in front of the screen, plus I can put the money elsewhere. So getting the widest dispersion possible out of only 2 speakers is always the goal for me. Speakers are currently 7.5 feet apart and I sit 9 feet from them in a 10.5' wide room.

 

[AJ Soundfield]

Well, if you can get them 3-4 ft. minimum away from boundaries. The X-Omni I posted were originally intended for surround, they just happen to image so well that the 2 channel community adopted them.

Dennis, that only applies to the mains positions aka "direct" frontal sound for our forward evolved ears. Surround/envelopment channels such as rears can be diffuse and decorrelated (in the case of music should be). Short reflection times from an omni closer to boundaries from 90 degrees on backwards is ok.

cheers,

AJ

 

[AJ Soundfield]

That was the argument that adherents claimed, but the reality is that the peaks and dips, such as from SBIR and room modes will exist regardless. The radiation from these speakers can shift them around a little, but it doesn't eliminate them. Earl Geddes and Linkwitz got into a friendly disagreement about this as it pertained to bass, so Geddes did a study comparing numerous bass radiation patterns and showed that it did not improve the bass response. While it did change how modes were excited, it didn't actually decrease deviance (meaning bass was no flatter or smoother, just different).

Matt,

I'm an AES member and cannot recall such a study, can you provide a link?
I'm also aware that there are those who subscribe to the belief that bass SQ is determined only by amplitude domain, as you are describing above. That a steady state signal and pressure points tells all. This might be true for HT. There is another view that the 3 particle velocities at those points are also important and that an impulsive and/or wavelet signal analysis is more appropriate for music oriented systems. This is the view held by Dr Griesinger JJohnston et al. I personally hold that view also, with the backing of this complied paper of 40+ bass studies in small room acoustics https://secure.aes.org/forum/pubs/conferences/?elib=17270
Hence me asking about Earls, thanks.

cheers,

AJ

 

[tesseract]

Dennis, that only applies to the mains positions aka "direct" frontal sound for our forward evolved ears. Surround/envelopment channels such as rears can be diffuse and decorrelated (in the case of music should be). Short reflection times from an omni closer to boundaries from 90 degrees on backwards is ok.

cheers,

AJ

I was wondering if the same applies to surrounds as fronts. Thanks for clearing that up, AJ!

 

[Matthew J Poes]

Matt,

I'm an AES member and cannot recall such a study, can you provide a link?
I'm also aware that there are those who subscribe to the belief that bass SQ is determined only by amplitude domain, as you are describing above. That a steady state signal and pressure points tells all. This might be true for HT. There is another view that the 3 particle velocities at those points are also important and that an impulsive and/or wavelet signal analysis is more appropriate for music oriented systems. This is the view held by Dr Griesinger JJohnston et al. I personally hold that view also, with the backing of this complied paper of 40+ bass studies in small room acoustics https://secure.aes.org/forum/pubs/conferences/?elib=17270
Hence me asking about Earls, thanks.

cheers,

AJ

He didn't publish it in JAES. It was part of a friendly disagreement with Linkwitz.
http://www.gedlee.com/Papers/rooms and sources_norm.pdf

I should take a look at what you are posting before commenting further. The view that I am ascribing to is that which has been published and discussed by Toole (and his Harman colleagues) as well as Geddes. All have published in AES as well as presented on the topic at AES conferences.

 

[Matthew J Poes]

Matt,

I'm an AES member and cannot recall such a study, can you provide a link?
I'm also aware that there are those who subscribe to the belief that bass SQ is determined only by amplitude domain, as you are describing above. That a steady state signal and pressure points tells all. This might be true for HT. There is another view that the 3 particle velocities at those points are also important and that an impulsive and/or wavelet signal analysis is more appropriate for music oriented systems. This is the view held by Dr Griesinger JJohnston et al. I personally hold that view also, with the backing of this complied paper of 40+ bass studies in small room acoustics https://secure.aes.org/forum/pubs/conferences/?elib=17270
Hence me asking about Earls, thanks.

cheers,

AJ

I think that study by Elias has some issues that concern me, given the small differences he found and differences in LF sources. I want to have an expert evaluate that study before I comment on it.

I can't find any reference to Wavelet analysis in David Griesinger's work, but his paper is interesting. The idea that LF sources can be spatially recognized would go against a lot of current thinking. I've had a lot of experts, with PhD's, who research this, including those studying HRTF, tell me that humans cannot detect LF spatial information under any condition, even free space.

My view would be if something is true in music reproduction it should be equally true in cinema reproduction, but..., if we accept some of what is being said here, it seems that some fundamental changes to how a cinema system is setup and deployed would be necessary.

In any case, you have raised my interest. I've looked into wavelet analysis before, specifically with regard to Legacy's work with it and the Wavelet processor. I've not made enough progress on the topic to feel very knowledgeable.

I asked Earl Geddes to comment on the study and he did get back to me. He said that the modulated transfer function has its adherents (and publications) but he is unware of any studies that showed there was any correlation with perceived sound. Suggesting its measuring something we cant hear.

He too is dismissive of that study from Elias. He should have tested more positions, multiple LF sources, and then seen what the results were. He also should have used the same driver in the two different enclosure types, as it's possible that with modulated transfer function small differences in the driver could show up.


I'm still skeptical but intrigued, so if you have anything more to share that isn't from AES (I can't download their articles, I don't have access).

 

[AJ Soundfield]

I'm still skeptical but intrigued, so if you have anything more to share that isn't from AES (I can't download their articles, I don't have access).

Sorry, pm'd you before reading this. Shame! Basic AES membership is $150 yr (?) and now gives full access to e-library, every paper ever published! Worth it's weight in gold.
That paper is critical to the entire subject. Feel free to ask Earl about ;-)