• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Diffusing 1st reflections of speakers that measure great on and off-axis - instead of absorbing

Duke

Major Contributor
Audio Company
Forum Donor
Joined
Apr 22, 2016
Messages
1,523
Likes
3,745
Location
Princeton, Texas
Sorry, I mean a soundfield mike. Capture both all 3 volume velocities and the pressure at two points about 15-20 CM apart, on a line parallel to the line between the speakers.

Remember that in any point in the atmosphere, there are 4 variables, 3 volume velocities (which can be represented in two ways, but velocity is easier to interpret than polar for what I'd care about) and the pressure. The pressure at one point (that's what an omni does) is only 1 of 4 variables at that point.

With all four variables you learn a great deal about what reflects where, what room modes look like, what various time delays are, and so on.

Thank you for replying J_J, and apologies for my slow reply. My little company is a one-man show and when something goes wrong that one man has to work a lot of unpaid overtime.

I looked into soundfield microphones and they look like an excellent tool for room acoustics evaluations.

I'm actually far more interested in optimizing loudspeakers for favorable speaker/room synergy. For example, regarding room acoustics you wrote that "dead is my preference." This implies that the less energy a speaker puts into reflections, the better. Which in turn implies that a highly directional speaker is preferable, especially for rooms where sufficiently aggressive acoustic treatment to render the room "dead" is impractical.
 

j_j

Major Contributor
Audio Luminary
Technical Expert
Joined
Oct 10, 2017
Messages
2,267
Likes
4,758
Location
My kitchen or my listening room.
No, I don't think "less energy into reflections" is the thing, rather "absorbing reflections" is the key. That way you can also use the speaker in a room with reflections and not sound strange. You'll never have directional speakers (aside from dipoles) at 50 or 100Hz in any practical fashion, so you're kidn of stuck with "flat direct, flat power response" in my view. While I'm not perhaps in complete agreement with Floyd, our disagreements are down in the details, as opposed to at the "fundamental" level.

The timbre of any actual reflections or dispersed reflections is very important in determining how things sound, and you can make things go very wrong, or quite right depending on how you achieve that.

I am not of the "no room no envelopment" idea, either. You can do surprising things with 2 speakers (and sound better in headphones at the same time with the same signal), but not with conventional production.

Production has not adapted to the accuracy of digital recording. Scrape flutter and the like have some quite remarkable effects.
 

Duke

Major Contributor
Audio Company
Forum Donor
Joined
Apr 22, 2016
Messages
1,523
Likes
3,745
Location
Princeton, Texas
No, I don't think "less energy into reflections" is the thing, rather "absorbing reflections" is the key. That way you can also use the speaker in a room with reflections and not sound strange.

I understand what you are saying, but am not convinced that speakers with fairly narrow and well-behaved radiation patterns "sound strange" in rooms with reflections.

You'll never have directional speakers (aside from dipoles) at 50 or 100Hz in any practical fashion, so you're kind of stuck with "flat direct, flat power response" in my view.

Dipoles and cardiods can do it, at the expense of requiring more power for a given low-end extension and/or SPL from a comparable-sized enclosure. The Gradient Revolution uses cardioid midbass and dipole bass, arguably to good effect. The Dutch & Dutch 8c uses cardioid midbass loading down to 100 Hz, then monopole bass below that, again arguably to good effect.

I'm not necessarily advocating for maintaining narrow directivity down low, but some imo very good designers do it.

The timbre of any actual reflections or dispersed reflections is very important in determining how things sound, and you can make things go very wrong, or quite right depending on how you achieve that.

I agree with this, and imo the arrival times and even the arrival directions of those (timbrally correct) reflections matter.
 

March Audio

Master Contributor
Audio Company
Joined
Mar 1, 2016
Messages
6,378
Likes
9,317
Location
Albany Western Australia
I understand what you are saying, but am not convinced that speakers with fairly narrow and well-behaved radiation patterns "sound strange" in rooms with reflections.



Dipoles and cardiods can do it, at the expense of requiring more power for a given low-end extension and/or SPL from a comparable-sized enclosure. The Gradient Revolution uses cardioid midbass and dipole bass, arguably to good effect. The Dutch & Dutch 8c uses cardioid midbass loading down to 100 Hz, then monopole bass below that, again arguably to good effect.

I'm not necessarily advocating for maintaining narrow directivity down low, but some imo very good designers do it.



I agree with this, and imo the arrival times and even the arrival directions of those (timbrally correct) reflections matter.
I hear what you are saying however being cardioid or directional (as far as is practically achievable) is still unlikely to stop much of the reflections that are happening in a domestic room.

As jj says the timbre of the reflections is important so in my view you are stuck with sorting the room.

I also prefer more "dead" rooms. I find "live" rooms confusing, fatiguing and mask detail.
 
Last edited:

youngho

Senior Member
Joined
Apr 21, 2019
Messages
486
Likes
799
No, I don't think "less energy into reflections" is the thing, rather "absorbing reflections" is the key. That way you can also use the speaker in a room with reflections and not sound strange. You'll never have directional speakers (aside from dipoles) at 50 or 100Hz in any practical fashion, so you're kidn of stuck with "flat direct, flat power response" in my view. While I'm not perhaps in complete agreement with Floyd, our disagreements are down in the details, as opposed to at the "fundamental" level.

The timbre of any actual reflections or dispersed reflections is very important in determining how things sound, and you can make things go very wrong, or quite right depending on how you achieve that.

I am not of the "no room no envelopment" idea, either. You can do surprising things with 2 speakers (and sound better in headphones at the same time with the same signal), but not with conventional production.

Production has not adapted to the accuracy of digital recording. Scrape flutter and the like have some quite remarkable effects.

@j_j when I read your posts, I always feel like I'm many steps behind. I think my own personal concern regarding absorbing reflections is that absorption of reflections is highly dependent on the angle, material, and fabric covering, which will likely modify the frequency spectrum of the reflections. No non-dipole speakers are directional down to 50-100 Hz, but it also seems like most non-dipole speakers are not flat into this frequency range in terms of their direct and power response, at least in terms of measurements.

How would you go about absorbing or dispersing reflections in a "quite right" way in terms of timbre?

Can you please expound on what you mean about "surprising things"?
 

j_j

Major Contributor
Audio Luminary
Technical Expert
Joined
Oct 10, 2017
Messages
2,267
Likes
4,758
Location
My kitchen or my listening room.
Youngho I wish there was a simple answer to all that, but there isn't.

You need to either know or measure the absorption of your material in bulk form, know the first angles of incidence, know the speaker response (both power and direct), and match it to 1) the room's absorption, 2) the room's "pressure response" point(s), and 3) whatever dispersion you might want, or not. Yes, this means a non-power-flat speaker can actually provide you with the right things, but you'll find that mostly impossible with things that beam too too much. It's also hard to do with a dipole or a bipole, because then the room is even more important. This is why I stick to specifics most of the time. I've found the 1960's and 1970's "rules of thumb" mostly being "as clumsy as thumbs". Yes, I'm well aware some people differ with that opinion.
 

youngho

Senior Member
Joined
Apr 21, 2019
Messages
486
Likes
799
@Bjorn and @localhost127, thank you for your edifying posts. I learned a lot about your approaches, but I have a number of questions that I hope you would be so kind as to help me understand. I understand if you're otherwise occupied.

1. Could you please explain to an ignorant person like me how the use of a proper ISD-gap (pardon my terminology) "control room"-style arrangement might be expected to affect listener preference of loudspeakers, which is what Harman was investigating (with their two speaker shuffling rooms set up before 1998 using available methods and materials (I think BAD panels came onto the market after that, but I'm probably mistaken), from what I understand), and how would you go about predicting how that would translate to other acoustic environments like most "typical" home listening rooms for stereo and multichannel?
2. Could you possibly further explain more of the differences in setup between stereo and multichannel control rooms with an adequate ISD gap, also how do you typically go about that in a domestic setting? Earl Geddes once talked about a futon on the floor with a rug on top between the speakers and the listener, for example, but I imagine that this is not trivial for most consumers, also not to mount a futon on the ceiling.
3. Related to #2, would you be so kind as to demonstrate examples of multichannel listening rooms that you believe reflect excellent understanding of small room acoustics? Even better, how that could be achieved in a domestic setting? Possibly even better, measurements? I would love to have Blackbird Studio C in one of my spare rooms, but unfortunately, that is not realistic.
4. Could you please comment on the use of a secondary set of speakers producing a delayed signal to mimic the Haas kicker effect?

Thank you in advance for your time and consideration,

Young-Ho
 

youngho

Senior Member
Joined
Apr 21, 2019
Messages
486
Likes
799
Youngho I wish there was a simple answer to all that, but there isn't.

You need to either know or measure the absorption of your material in bulk form, know the first angles of incidence, know the speaker response (both power and direct), and match it to 1) the room's absorption, 2) the room's "pressure response" point(s), and 3) whatever dispersion you might want, or not. Yes, this means a non-power-flat speaker can actually provide you with the right things, but you'll find that mostly impossible with things that beam too too much. It's also hard to do with a dipole or a bipole, because then the room is even more important. This is why I stick to specifics most of the time. I've found the 1960's and 1970's "rules of thumb" mostly being "as clumsy as thumbs". Yes, I'm well aware some people differ with that opinion.

Thank you. As probably expected, the more I read, the less I realize that I can pretend to know. It seems simple to say "broadband absorption," but the density and intrinsic properties of the material, the spacing from the boundary, the angle of incidence, and the covering fabric all seem to matter, and relatively little seem to be well-reflected in any published data (regarding fabric, for example, GoM acoustical performance data for Anchorage https://www.guilfordofmaine.com/swatches/2334-2335-2839 versus http://www.acousticfrontiers.com/062019-newsletter/, as well as graphs published by Floyd Toole and provided by Peter D'Antonio). Similarly, dispersion for a hemicylindrical element versus array, various types of diffusion, as well as depths and setups/arrangements (direction, lobing, etc).

I found your comment about "mostly impossible with things that beam too too much" extremely interesting. Could I trouble you to explain this one specific point further? For example, I wonder how the so-called Shirley curve affects perceptions of tonality with respect to speaker radiation patterns and room absorption, but probably this is not at all what you're getting at.

Thanks so much,

Young-Ho
 

patate91

Active Member
Joined
Apr 14, 2019
Messages
253
Likes
137
Thank you. As probably expected, the more I read, the less I realize that I can pretend to know. It seems simple to say "broadband absorption," but the density and intrinsic properties of the material, the spacing from the boundary, the angle of incidence, and the covering fabric all seem to matter, and relatively little seem to be well-reflected in any published data (regarding fabric, for example, GoM acoustical performance data for Anchorage https://www.guilfordofmaine.com/swatches/2334-2335-2839 versus http://www.acousticfrontiers.com/062019-newsletter/, as well as graphs published by Floyd Toole and provided by Peter D'Antonio). Similarly, dispersion for a hemicylindrical element versus array, various types of diffusion, as well as depths and setups/arrangements (direction, lobing, etc).

I found your comment about "mostly impossible with things that beam too too much" extremely interesting. Could I trouble you to explain this one specific point further? For example, I wonder how the so-called Shirley curve affects perceptions of tonality with respect to speaker radiation patterns and room absorption, but probably this is not at all what you're getting at.

Thanks so much,

Young-Ho

You can look at what Ethan Winer wrote about that. Note that toi can reach him over his forum.
 

Duke

Major Contributor
Audio Company
Forum Donor
Joined
Apr 22, 2016
Messages
1,523
Likes
3,745
Location
Princeton, Texas
I hear what you are saying however being cardioid or directional (as far as is practically achievable) is still unlikely to stop much of the reflections that are happening in a domestic room.

As part of a controlled-directivity loudspeaker system, cardioids can reduce the amount of energy that goes into some reflections at some frequencies. Whether or not it's "worth the trouble" would be another topic.

As jj says the timbre of the reflections is important so in my view you are stuck with sorting the room.

Well yes, I'm just in favor of making that task a bit easier.

I also prefer more "dead" rooms. I find "live" rooms confusing, fatiguing and mask detail.

Is it possible that what you actually dislike is having a lot of energy in the early reflections?

"The earlier a reflection arrives the more it contributes to masking the direct sound." - David Griesinger

"When presence is lacking the earliest reflections are the most responsible." - David Griesinger

"The earlier and the greater in level the first room reflections are, the worse they are... While it is certainly true that all reflections add to spaciousness, the very early ones (< 10 ms.) do so at the sake of imaging and coloration... The first reflections in small rooms must be thought of as a serious problem that causes coloration and image blurring. These reflections must be considered in the loudspeaker design and should be also be considered in the room as well." - Earl Geddes
 

Bjorn

Major Contributor
Audio Company
Forum Donor
Joined
Feb 22, 2017
Messages
1,286
Likes
2,562
Location
Norway
50-100 Hz is modal area and not specular. Directivity of a speaker in this frequency range is swamped by room modes in small rooms.
 

March Audio

Master Contributor
Audio Company
Joined
Mar 1, 2016
Messages
6,378
Likes
9,317
Location
Albany Western Australia
As part of a controlled-directivity loudspeaker system, cardioids can reduce the amount of energy that goes into some reflections at some frequencies. Whether or not it's "worth the trouble" would be another topic.



Well yes, I'm just in favor of making that task a bit easier.



Is it possible that what you actually dislike is having a lot of energy in the early reflections?

"The earlier a reflection arrives the more it contributes to masking the direct sound." - David Griesinger

"When presence is lacking the earliest reflections are the most responsible." - David Griesinger

"The earlier and the greater in level the first room reflections are, the worse they are... While it is certainly true that all reflections add to spaciousness, the very early ones (< 10 ms.) do so at the sake of imaging and coloration... The first reflections in small rooms must be thought of as a serious problem that causes coloration and image blurring. These reflections must be considered in the loudspeaker design and should be also be considered in the room as well." - Earl Geddes

Quite possible :)
 

Bjorn

Major Contributor
Audio Company
Forum Donor
Joined
Feb 22, 2017
Messages
1,286
Likes
2,562
Location
Norway

j_j

Major Contributor
Audio Luminary
Technical Expert
Joined
Oct 10, 2017
Messages
2,267
Likes
4,758
Location
My kitchen or my listening room.
"The earlier a reflection arrives the more it contributes to masking the direct sound." - David Griesinger

"When presence is lacking the earliest reflections are the most responsible." - David Griesinger

"The earlier and the greater in level the first room reflections are, the worse they are... While it is certainly true that all reflections add to spaciousness, the very early ones (< 10 ms.) do so at the sake of imaging and coloration... The first reflections in small rooms must be thought of as a serious problem that causes coloration and image blurring. These reflections must be considered in the loudspeaker design and should be also be considered in the room as well." - Earl Geddes

Nothing to argue with there, especially for seriously specular reflections.

On the other hand, a truly dead room with very well matched speakers, can quite seriously give you the "central image in your head" effect, too.

Most speakers aren't that well matched. Some are. :)
 

dasdoing

Major Contributor
Joined
May 20, 2020
Messages
4,208
Likes
2,673
Location
Salvador-Bahia-Brasil

youngho

Senior Member
Joined
Apr 21, 2019
Messages
486
Likes
799
that's a non-issue. you won't hear reflections below -20dB. see 3.4.1 and 3.4.2 i this paper: https://users.aalto.fi/~ktlokki/Publs/mst_laukkanen.pdf
keeping ETC below -20dB is not hard at all. since you mentioned fabric; if you can blow through, highs can go through

I don't think it's that trivial: https://www.audiosciencereview.com/...f-axis-instead-of-absorbing.14096/post-471238. It's not like all reflections would be uniformly attenuated -20dB but rather certain parts of their frequency spectra would be more attenuated than others, also depending on the angle of incidence. Look at the graphs, then come back about "-20 dB" for the highs.

Young-Ho
 

youngho

Senior Member
Joined
Apr 21, 2019
Messages
486
Likes
799
I don't have time to go into all our questions, but this one is something I touched on in my previous post (see last paragraph):
https://www.audiosciencereview.com/...nstead-of-absorbing.14096/page-13#post-474136

Adding more sources is radical different from a Haas kicker and a strong termination of the ISD gap.

Ah, Duke and Brian Waslo use a different approach than you touched on in your post. You didn't actually address this approach, which is why I brought it up. The delayed signal comes from two channels placed very close to the main left and right speakers, which in both Duke's and Brian Waslo's approaches involved highly controlled directivity main channels. THESE ARE NOT ACTIVE SURROUNDS. I would let @Duke address this further, rather than myself.

Thanks,

Young-Ho
 

youngho

Senior Member
Joined
Apr 21, 2019
Messages
486
Likes
799
On the other hand, a truly dead room with very well matched speakers, can quite seriously give you the "central image in your head" effect, too.
Most speakers aren't that well matched. Some are. :)

@j_j, do you consider the "central image in your head" generally desirable as a goal, in your view?
 

Duke

Major Contributor
Audio Company
Forum Donor
Joined
Apr 22, 2016
Messages
1,523
Likes
3,745
Location
Princeton, Texas
The delayed signal comes from two channels placed very close to the main left and right speakers, which in both Duke's and Brian Waslo's approaches involved highly controlled directivity main channels. THESE ARE NOT ACTIVE SURROUNDS. I would let @Duke address this further, rather than myself.

I'm not familiar with Brian Waslo's approach, but briefly what I do is, use a fairly directional rear-firing array to increase the amount of energy in relatively late-onset reflections without correspondingly increasing the energy in early (within 10 milliseconds of the direct sound) reflections.
 
Last edited:

youngho

Senior Member
Joined
Apr 21, 2019
Messages
486
Likes
799
Top Bottom