A Broad Discussion of Speakers with Major Audio Luminaries

[Cote Dazur]

The insistance of many here in trying to project multi channel as a de facto music playing solution for enjoying stereo music at home never cease to make me wonder.
A stereo system to play stereo music at home has so many advantages.
For movies, particularly the action pack American productions, which is where multi channel has its roots, if you are into that, then yes, thunderous bass and speaker galore, not to miss any noise from that helicopter coming from behind the listener is a very valid option.
For most, dare I say interesting music, 2 speakers, adequately positioned in a room prepared for music reproduction is the elegant and mature solution.
After all, last I checked the overwhelming majority of music we already posses or is offered was produced in stereo.

 

[kyuu]

The insistance of many here in trying to project multi Chanel as a de facto music playing solution for enjoying stereo music at home never cease to make me wonder.

What are you even talking about? Are you talking about upmixing? I'm one of those who upmixes pretty much everything if so, and I wonder at anyone who doesn't personally. Actual multichannel mixes are generally (though not always) superior, of course.

A stereo system to play stereo music at home has so many advantages.

Beyond only needing two speakers, I don't see any.

For movies, particularly the action pack American productions, which is where multi channel has its roots, if you are into that, then yes, thunderous bass and speaker galore, not to miss any noise from that helicopter coming from behind the listener is a very valid option.

Multichannel has little to do with "thunderous bass", except for the LFE channel (but you don't need an LFE channel for loud bass). There's obviously a lot more to it than just helicopter noises. Don't be reductive.

For most, dare I say interesting music, 2 speakers, adequately positioned in a room prepared for music reproduction is the elegant and mature solution.

The hell does "elegant and mature" mean? 2 speakers is adequate, but there's nothing ideal about it in terms of music reproduction. It didn't become the de facto standard for music reproduction because it was technically superior.

After all, last I checked the overwhelming majority of music we already posses or is offered was produced in stereo.

Again, the fact that it's the de facto standard has nothing to do with its technical merits.

If you prefer to stick with two speakers, that's fine. But you're not providing any sort of real argument as to why it's superior to multichannel.

 

[BookshelfBob]

The insistance of many here in trying to project multi channel as a de facto music playing solution for enjoying stereo music at home never cease to make me wonder.
A stereo system to play stereo music at home has so many advantages.
For movies, particularly the action pack American productions, which is where multi channel has its roots, if you are into that, then yes, thunderous bass and speaker galore, not to miss any noise from that helicopter coming from behind the listener is a very valid option.
For most, dare I say interesting music, 2 speakers, adequately positioned in a room prepared for music reproduction is the elegant and mature solution.
After all, last I checked the overwhelming majority of music we already posses or is offered was produced in stereo.

There is all sorts of evidence 5.1 is superior for movies. Past that you hit pretty heavy diminishing returns. A good 5 channel mix will always beat the stereo for a movie. Irrelevant of how poor the nation making it is.

And if mixed properly. Anything going for a "live" feel in music is going to sound massively more immersive in a multi channel scenario.

 

[Mort]

For most, dare I say interesting music, 2 speakers, adequately positioned in a room prepared for music reproduction is the elegant and mature solution.

I'm not sure you could get three people on this site to agree what that even means much less persuaded by those arguments.

 

[Cote Dazur]

I'm not sure you could get three people on this site to agree what that even means much less persuaded by those arguments.

You might be right, but I very much doubt it, I think I can go up to 4 or 5.
I am not trying to persuade any body, as far as I am concerned any one can do what he believes is best.
Seems from the few post above the I am correct when saying “The insistance of many here in trying to project multi channel as a de facto music playing solution for enjoying stereo music at home”
Here, meaning ASR, outside, in the wild, 2 speaker stereo, still is the norm.
Just saying that I prefer stereo, not saying Multi is not great, just not great for me and not “necessary” to enjoy stereo at home.

 

[RexrothPigeon]

I am not trying to persuade any body, as far as I am concerned any one can do what he believes is best.

lol your post was an argument intended to persuade, otherwise you would have just muttered it to yourself under your breath and none of us would have to see it.

 

[hex168]

Several months ago I had a brief but interesting conversation with Edgar Choueri, inventor of the BACCH system, in which I asked him about in-room reflection arrival times. He said that reflections arriving later than 20 millisecond are not detrimental to crosstalk cancellation. I seem to recall @Bjorn also recommending 20 milliseconds delay before the strong onset of reflections. Unfortunately that much delay is probably only practical for a few.

@Bjorn, what are your thoughts on hemicylindrical diffusors? Seems to me they meet your criteria but have the downside of taking up a lot of space in the room.

You should be able to get 20 msec delay in a reasonably-sized room by placing horn tweeters directly in front of the listener (think coffee table or ottoman) and facing the front wall, at least for frequencies for which the horn maintains control of directivity. If I am interpreting Griesinger correctly, this is most important above 700 Hz anyway, so it should be possible to do that.

 

[NTK]

I created a Mathematica notebook for the sidewall reflection calculation. If I have the system, listener and 2 speakers, setup as a 12 ft equilateral triangle, in order to have a 10 ms delay for the lateral reflection, the speaker needs to be 7.4 ft (dimension "r") from the wall. Not that easy.

 

[Duke]

You should be able to get 20 msec delay in a reasonably-sized room by placing horn tweeters directly in front of the listener (think coffee table or ottoman) and facing the front wall, at least for frequencies for which the horn maintains control of directivity. If I am interpreting Griesinger correctly, this is most important above 700 Hz anyway, so it should be possible to do that.

I'm having a hard time envisioning what you describe, but I do agree with the idea that good directional control combined with well thought-out aiming can push those first lateral reflections considerably further back in time than would normally have been the case. And imo good pattern control down to 700 Hz would be great whether or not you can get that 20 msec initial time delay gap.

I created a Mathematica notebook for the sidewall reflection calculation. If I have the system, listener and 2 speakers, setup as a 12 ft equilateral triangle, in order to have a 10 ms delay for the lateral reflection, the speaker needs to be 7.4 ft (dimension "r") from the wall. Not that easy.

View attachment 555057

If you start out with speakers having a radiation pattern width of 90 degrees or less (45 degrees or less to either side of the centerline), and then you toe them in aggressively such that their axes criss-cross in front of the listening position (so maybe 45 degrees of toe-in), you effectively avoid having a strong first ipsilateral reflection with its short reflection path. The first strong lateral reflection becomes the contralateral one, which will probably arrive after 10 milliseconds initial time delay gap. The reflections off the wall behind the listener still need to be addressed, but imo progress will have been made.

 

[Vuki]

I'm not sure you could get three people on this site to agree what that even means much less persuaded by those arguments.

I'm one of three...

 

[Bjorn]

My understanding is that the basic principle of having a fairly long time gap between the direct sound and the first significant early horizontal-plane reflections in the studio control room is not obsolete, but that there are now better ways of getting there, such as the Reflection-Free Zone approach.

The RFZ stems from LEDE. Creating the initial time-delay gap is essentially what the reflection-free zone is about. The goal was to extend it by 2–5 ms beyond the ITD of the recording room, in order to hear the recorded signal as accurately as possible. This was either achieved through carefully controlled absorption or by redirecting the energy with slanted walls or panels. The latter has several advantages and can also be applied to the ceiling.

And this part of the LEDE design has been incorporated into the ITU-R BS.1116, EBU Tech 3276 and AES TD1001 standards, where the requirement is to attenuate the first 15 ms to a minimum of -10 dB. Of course, the LEDE design was much more stringent here, with reflections attenuated to a minimum of -20 dB, and preferably to -25 or -30 dB, making the reflection-free zone effectively anechoic.

The LEDE concept of retaining late diffuse energy is also still very much used in control rooms today. Not always with an ISD gap 2–5 ms longer than that of the recording room, but with a diffuse rear wall providing generally late-arriving energy. I doubt anyone has complete data on what is being done today, but I see this design principle being followed more often than other approaches, such as non-environmental rooms à la Hidley or Northward's FTB design. In my opinion, Northward's Front-to-Back (FTB) rooms are a clear step away from accuracy.

The longer that initial time-delay gap, the larger the perceived acoustic space, as @Bjorn mentions:



In other words, if you can push the first significant reflections further back in time than they would normally be in your room, you can make it sound like your room is bigger than it actually is.

I think there is an important nuance here. The objective of LEDE was not simply to make the initial time-delay gap as long as possible. The requirement was that the control room's ISD should be longer than the ITD of the recording space, by at least roughly 2–5 ms. The idea was to allow the listener to hear the spatial information from the recording space before the control room's own reflections became significant.

So I wouldn't interpret it as “the longer the ITD gap, the larger the perceived acoustic space.” There was no goal of maximizing the RFZ or ISD indefinitely. The relationship to the recording space was the important part.

The roughly 20 ms figure is nevertheless very interesting. Davis connected it with Beranek's work on the best concert halls and with the Haas region, and this seems to have made approximately 20 ms a particularly attractive target when the room permitted it. But I see this more as a useful and perceptually significant region, rather than a rule that increasing the gap indefinitely will continue to make the room sound larger.

In fact, Davis described a fairly wide range of possible ISDs depending on the room. So I think the more accurate statement would be: the control room should have an initial time-delay gap sufficiently longer than that of the recording space, with a value around 20 ms being particularly interesting because of its relationship to concert-hall acoustics and the Haas effect.

 

[MaxwellsEq]

The idea was to allow the listener to hear the spatial information from the recording space before the control room's own reflections became significant.

I think all the historical schemes and current standards are based on this requirement.

The control room and its monitoring system is supposed to be an analytical tool. I think it helps to consider that it exists to identify "faults" in the microphones or recorded space, such as, buzzes, squeaks and unwanted echoes or resonances. If the control room is too echoe-y it will mask faults in the recorded space.

 

[Bjorn]

As mentioned, highly directive loudspeakers can also be used to achieve a reflection-free zone with little or no absorption, depending on the frequency at which their dispersion starts to widen. This is very similar to my own setup in the living room upstairs. I only use two absorption panels, yet the early reflections are still highly attenuated. The rear wall is covered with diffusers.

My speakers here are large horn speakers with narrow and very constant directivity down to low frequencies. A similar approach can also be achieved with planar speakers, although most of them are dipoles and therefore radiate substantially towards the rear, while also becoming quite directional at higher frequencies. And clearly, they don't have the same dynamics as horn-loaded speakers.

 

[patoulol]

I think there s an important distinction between playing stereo recordings and saying that stereo is the only really valid way of listening to music

a good twochannel system, with properly positioned speakers and a well controlled room, can obviously sound fantastic. I wouldnt argue with that at all

but native multichanel or immersive mixes are a different thing.

The extra channels arent necessarily there just to add more effects or make things more spectacular
They give the mixer more freedom to place and control spatial information.

I work with Dolby Atmos Music, and what I find interesting is that a good Atmos mix doesn’t necessarily sound like “more speakers”

It can actually feel very natural, because the spatial information is part of the mix itself rather than being created mainly through the interaction of two speakers with the listening room

that doesn’t mean stereo is obsolete, or that everyone needs multichannel at home. If someone prefers two-channel, that’s perfectly reasonable

I just don’t think one format needs to be presented as inherently more “mature” than the other

in the end, the quality of the speakers, their directivity, their positioning and the room still matter enormously

whatever the number of channel

 

[RickS]

My understanding is that the basic principle of having a fairly long time gap between the direct sound and the first significant early horizontal-plane reflections in the studio control room is not obsolete, but that there are now better ways of getting there, such as the Reflection-Free Zone approach.

The longer that initial time-delay gap, the larger the perceived acoustic space, as @Bjorn mentions:



In other words, if you can push the first significant reflections further back in time than they would normally be in your room, you can make it sound like your room is bigger than it actually is.

Thanks, my bad for relying on AI to cross check. Based on a previous post, I had asked Gemini whether LEDE was used in contemporary recording studios. The response was No, that it is obsolete. Asked again today and the answer has changed to Yes! Gemini must have used limited or incomplete data for its original conclusion. In any case, clearly the AI response is not very reliable.

P.S.

While the initial Gemini response may have been overly negative on LEDE, it did mention that RFZ was the successive approach.

 

[Newman]

The insistance of many here in trying to project multi channel as a de facto music playing solution for enjoying stereo music at home never cease to make me wonder.

The only insistence I see is that experiments have shown that (for stereo recordings, per your insistence) good upmixing has the potential to improve perceptions of ASW and ENV, which is proven to correspond with perceptual preference, while not simultaneously degrading soundstage imaging, which also corresponds with perceptual preference.

A stereo system to play stereo music at home has so many advantages.

Stereo

Stereo

Stereo

Let me guess the 'so many advantages': cheap, compact unobtrusive? Maybe, but only if you exclude 'serious audiophiles'.

For movies, particularly the action pack American productions, which is where multi channel has its roots, if you are into that, then yes, thunderous bass and speaker galore, not to miss any noise from that helicopter coming from behind the listener is a very valid option.

This is the sort of comment that belongs in the 1990s Letters to the Editor section of Stereophile magazine.

We should not still be stuck in such paradigms.

For most, dare I say interesting music, 2 speakers, adequately positioned in a room prepared for music reproduction is the elegant and mature solution.

Only if you are happy with less than the best sonic rewards, for a given budget.

After all, last I checked the overwhelming majority of music we already posses or is offered was produced in stereo.

Means nothing. MCH systems can still work in stereo any time the owner wishes to do so. Then they can offer so much more when the owner finally overcomes his or her Late Adopter Syndrome and tries upmixing stereo or obtaining discrete MCH recordings.

 

[hex168]

I'm having a hard time envisioning what you describe, but I do agree with the idea that good directional control combined with well thought-out aiming can push those first lateral reflections considerably further back in time than would normally have been the case. And imo good pattern control down to 700 Hz would be great whether or not you can get that 20 msec initial time delay gap.



If you start out with speakers having a radiation pattern width of 90 degrees or less (45 degrees or less to either side of the centerline), and then you toe them in aggressively such that their axes criss-cross in front of the listening position (so maybe 45 degrees of toe-in), you effectively avoid having a strong first ipsilateral reflection with its short reflection path. The first strong lateral reflection becomes the contralateral one, which will probably arrive after 10 milliseconds initial time delay gap. The reflections off the wall behind the listener still need to be addressed, but imo progress will have been made.

I did not write clearly enough, mistaking brevity for clarity; I'll try again:
A system could consist of controlled directivity front speakers, augmented with separate late-reflection generating speakers. The main speakers are in standard position, near the front wall, facing the listener. The front wall is defined as the wall immediately behind the main speakers. The augmenting speakers are in separate boxes, positioned immediately in front of the listener (or possibly to the side if directional control is tight enough). The augmenting speakers consist of a waveguide maintaining directional control down to 700 Hz (example ZXPC 18" x 10") loaded with either a 2" compression driver (expensive) or a 2" full-range cone driver (less so) and are high passed at 700Hz. The size of the horn can be reduced at the cost of raising the high pass frequency. The augmenting speakers are aimed away from the listener generally towards the front speakers, possibly offset towards the room corners for a longer path length. The path length would be: from the augmented speaker position near the listener, past the front speakers to the front wall corners and back to the listener. If the distance from the main speaker to the listener is 10 feet, than the total path length from the augmenting speakers would be greater than 10 feet plus 10 feet, or over 20 msec without using any added delay. To further clarify, the augmenting speaker is positioned such that the direct sound is largely inaudible from the listening position, many dB down from the reflection from the front wall. If necessary to reduce the direct sound from the augmenting speaker to the listener, the mouth of the horn could be surrounded by absorbing material.

 

[Duke]

I think there is an important nuance here. The objective of LEDE was not simply to make the initial time-delay gap as long as possible. The requirement was that the control room's ISD should be longer than the ITD of the recording space, by at least roughly 2–5 ms. The idea was to allow the listener to hear the spatial information from the recording space before the control room's own reflections became significant. (emphasis Duke's)

Thank you for the in-depth and educational reply, and yes ime one big perceptual benefit of increasing the ITDG in a home audio setting is pretty much the same as what you describe in studios.

So I wouldn't interpret it as “the longer the ITD gap, the larger the perceived acoustic space.” There was no goal of maximizing the RFZ or ISD indefinitely. The relationship to the recording space was the important part.

The roughly 20 ms figure is nevertheless very interesting. Davis connected it with Beranek's work on the best concert halls and with the Haas region, and this seems to have made approximately 20 ms a particularly attractive target when the room permitted it. But I see this more as a useful and perceptually significant region, rather than a rule that increasing the gap indefinitely will continue to make the room sound larger.

In fact, Davis described a fairly wide range of possible ISDs depending on the room. So I think the more accurate statement would be: the control room should have an initial time-delay gap sufficiently longer than that of the recording space, with a value around 20 ms being particularly interesting because of its relationship to concert-hall acoustics and the Haas effect.

Yeah, that was "painting with a broad brush" on my part. Thanks for fixing it.

As mentioned, highly directive loudspeakers can also be used to achieve a reflection-free zone with little or no absorption, depending on the frequency at which their dispersion starts to widen. This is very similar to my own setup in the living room upstairs. I only use two absorption panels, yet the early reflections are still highly attenuated. The rear wall is covered with diffusers.

My speakers here are large horn speakers with narrow and very constant directivity down to low frequencies. A similar approach can also be achieved with planar speakers, although most of them are dipoles and therefore radiate substantially towards the rear, while also becoming quite directional at higher frequencies. And clearly, they don't have the same dynamics as horn-loaded speakers.

I've never had a room where 20 milliseconds ITDG was feasible, nor speakers with directivity control down as low as yours, but efforts in that direction have still paid dividends as far as spatial quality goes. Imo. That being said, I'm sure my space could benefit from good diffusors.

 

[JPA]

I have a multichannel question that I think isn't very far off-topic. Maybe Floyd knows the answer.

If you have a multichannel system with front-wide speakers, does the surround processor synthesize (create) first lateral reflections from the L & R channels and feed those synthesized reflections to the front-wide speakers? If the answer is yes, then clearly we would want to absorb the first side reflections from the L & R speakers in order to not interfere with the synthesized reflections. If the answer is no, then we're back to deciding whether or not to absorb the 'real' first reflections based on room size, personal preference, etc.

 

[Duke]

A system could consist of controlled directivity front speakers, augmented with separate late-reflection generating speakers.

Yes!!! I've been doing this in various forms since 2013. My colleagues and I call the separate late-reflection generating speakers "LORAs", which stands for "Late Onset Reflection Assist." The first generation version fired up at the ceiling to get path length; we still use the vertical dimension but with some angling in the horizontal plane as well.

How did you come up with this idea?

Are you doing this in your system?

The main speakers are in standard position, near the front wall, facing the listener. The front wall is defined as the wall immediately behind the main speakers. The augmenting speakers are in separate boxes, positioned immediately in front of the listener (or possibly to the side if directional control is tight enough). The augmenting speakers consist of a waveguide maintaining directional control down to 700 Hz (example ZXPC 18" x 10") loaded with either a 2" compression driver (expensive) or a 2" full-range cone driver (less so) and are high passed at 700Hz. The size of the horn can be reduced at the cost of raising the high pass frequency. The augmenting speakers are aimed away from the listener generally towards the front speakers, possibly offset towards the room corners for a longer path length.

I've been using a different set-up geometry and smaller horns that don't go as low, with the late-reflection generating speakers behind the main speakers such that the main speakers "shield" them from line-of-sight.

I can see how the setup geometry you describe maximizes the reflection path length of the late-reflection generating speakers.

The path length would be: from the augmented speaker position near the listener, past the front speakers to the front wall corners and back to the listener. If the distance from the main speaker to the listener is 10 feet, than the total path length from the augmenting speakers would be greater than 10 feet plus 10 feet, or over 20 msec without using any added delay.

It's the time delay gap between the arrival on the direct sound and the arrival of the additional reflection energy that matters, so (unless I misunderstood you) the ten feet distance for the direct sound would have to be subtracted from the total path length of the additional reflections to arrive at the time delay gap.

To further clarify, the augmenting speaker is positioned such that the direct sound is largely inaudible from the listening position, many dB down from the reflection from the front wall. If necessary to reduce the direct sound from the augmenting speaker to the listener, the mouth of the horn could be surrounded by absorbing material.

Yes, that's why I like to use the main speakers to block any direct sound leakage from the late-reflection generating speakers.

Ime with a good initial set-up this approach has both spatial quality and sound quality benefits. The sound quality benefit arises from using the additional reflection energy to deliberately correct the in-room reflection field's spectral balance.