• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required as is 20 years of participation in forums (not all true). Come here to have fun, be ready to be teased and not take online life too seriously. We now measure and review equipment for free! Click here for details.

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

Music1969

Major Contributor
Joined
Feb 19, 2018
Messages
1,043
Likes
325
#1
Going back to 1970s style of mixing/master studios where all reflections were considered to be "bad," people have built some kind of online consensus that you need to go and "find all first reflections and absorb them." This is counter to research that indicates that side reflections can be beneficial. And that if you put too much absorption in a room, it can be come quite "dead" and dull sounding. With well designed speakers, reflections have similar tonality to direct sound and help reinforce that sound. This improves such things as speech intelligibility.
Hi @amirm (also @j_j @mitchco )

Putting aside room bass issues (<500 Hz), if one has really good on and off-axis measurements, like you mentioned from these JBL's and Genelec 8341A and others:

The on and off-axis sounds are very similar to each other as indicated by the blue dashed line being rather smooth. This means the speaker is room friendly as reflections sound similar to direct sound of the speaker. And gives you flexibility to position the speaker angle as you like.
Question: Non-linear absorbers could change the tone of 1st reflections of great speakers like these - but are there any downsides to diffusing these 1st reflections, especially linear diffusers like:

1592203309347.png
 

RayDunzl

Major Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
9,483
Likes
7,622
Location
Riverview FL
#2
Your graphic seems to be missing something.
 
OP
M

Music1969

Major Contributor
Joined
Feb 19, 2018
Messages
1,043
Likes
325
Thread Starter #3
Your graphic seems to be missing something.
LOL i cut off the bottom curve because that's absorption but yes that means I cut off the x-axis. See below.

But for purposes of this discussion I don't want to focus too much on this particular model of diffuser but assume we are talking about one that extends like this to 10kHz? Or higher?

So I'm just asking for any disadvantages of diffusing the 1st reflections of great speakers. We've seen discussion of using non-linear absorbers but linear diffusers seem easier to find (or make)

1592204107023.png
 
Last edited:

j_j

Addicted to Fun and Learning
Audio Luminary
Technical Expert
Joined
Oct 10, 2017
Messages
735
Likes
1,191
Location
My dining room.
#4
I'm not sure what you mean about non-linear absorption, did you mean not uniform across frequency?

If you did, then that can color the room response, but not the direct. This can work either for or against you. I'm curious how good diffusion happens at 100Hz, how big are these diffusors?

I'm just trying to make sure I understand what you mean, because actual nonlinearities are rather a troublesome lot of stuff.
 

pozz

Machine
Forum Donor
Editor
Joined
May 21, 2019
Messages
2,064
Likes
2,960
#7
Think of a diffusor as a more complicated flat wall. A wall is actually really effective in scattering reflections but, as you get farther away, starts producing more and more concentrated lobing, while true diffusors introduce time and amplitude variations to reflections over a wide range of angles. Absorbers work the same way, but only for amplitude (so no scattering or time variations are introduced).

Diffusor effectiveness ends us being determined by 1) incident angle, 2) receiver angle, 3) size, 4) distance.

When you look at that on a frequency-dependent basis, the wavelength of sound below 1kHz starts produce lobing issues in diffusors (in the same sense as you would talk about polar response and lobing in loudspeakers) so the pro companies like RPG limit diffusion around there, the bottom-end limit being 500Hz. I understand there are ways to design around this but it's not simple, requiring whole-room simulations, and usually avoided for small rooms.

Diffusors also absorb sound, which has to be taken into account, and the absorption pattern is not even.

So, net, there is no reason not to use diffusors at first-reflection points apart from 1) the colouration (i.e., frequency-specific effects) that others mentioned above and 2) the lack of distance from diffusor to the listening position. Trevor Cox and Peter D'Antonio recommend something like a minimum of 2m to 3m if I remember right, though the technical recommendation would be based on the wavelength of the lowest effective frequency.

That specific graph you posted looks like it's for the Visacoustic product. Consider it fanciful, heavily smoothed or representative of a hyper-specific use case. It is really hard to achieve that kind of even response.
 

Bjorn

Senior Member
Joined
Feb 22, 2017
Messages
388
Likes
404
Location
Norway
#9
The problem with diffusing "first reflections" in most small rooms is that there isn't sufficient distance for broadband diffusion. Therefore, the diffusion ends up being a bandlimited treatment which alters the sprectrum content to the degree of what's being used. That doesn't mean is still can't work ok and some may prefer it as well, but broadband treatment is always more correct or "neutral".

It's also a question of accuracy vs a more spacious sound field. Attenuating early arriving specular reflections gives the highes insight into the recorded material.

A middle way is using a hybrid product like RPG BAD Arc that diffuses primarily above 800 Hz and absorbs primarily below 800 Hz. This way one can achieve effect down to the Schroeder or even lower for that matter with no great distance required. However, it doesn't diffuse like a high quality 1D diffuser.
 
OP
M

Music1969

Major Contributor
Joined
Feb 19, 2018
Messages
1,043
Likes
325
Thread Starter #10
Thanks for all the replies!

The reason I'm interested in diffusing is after reading @j_j's old presentation attached.

I may be mis-interpreting very badly here (feel free to correct my misunderstanding) but it suggests that the brain is more sensitive to early arrivals (typically direct sound and 1st reflections ? ) and the brain is very good at filtering out late arrivals (2nd, 3rd, 4th etc reflections)?

So after reading this I guessed that good diffusing can be highly effective in making 1st reflections become later arrivals ?

One key assumption I made was that good (broadband) diffusing is easier to achieve than good broadband absorption but I now see this can be a bad assumption.
 

Attachments

Last edited:

j_j

Addicted to Fun and Learning
Audio Luminary
Technical Expert
Joined
Oct 10, 2017
Messages
735
Likes
1,191
Location
My dining room.
#11
Thanks for all the replies!

The reason I'm interested in diffusing is after reading @j_j's old presentation attached.

I may be mis-interpreting very badly here (feel free to correct my misunderstanding) but it suggests that the brain is more sensitive to early arrivals (typically direct sound and 1st reflections ? ) and the brain is very good at filtering out late arrivals (2nd, 3rd, 4th etc reflections)?

So after reading this I guessed that good diffusing can be highly effective in making 1st reflections become later arrivals ?

One key assumption I made was that good (broadband) diffusing is easier to achieve than good broadband absorption but I now see this can be a bad assumption.
In particular, signals after the direct up to a few milliseconds are very much suppressed, except as they add frequency shaping, and late stuff to maybe 20 milliseconds or so is somewhat suppressed as well. It's not so much the brain as the detectors on the basilar membrane for the early effects, later effects are in the CNS.

Whoa - no not MORE sensitive to early arrivals, LESS sensitive, except as frequency shaping. LESS sensitive.
 
Last edited:

Bjorn

Senior Member
Joined
Feb 22, 2017
Messages
388
Likes
404
Location
Norway
#12
Thanks for all the replies!

The reason I'm interested in diffusing is after reading @j_j's old presentation attached.

I may be mis-interpreting very badly here (feel free to correct my misunderstanding) but it suggests that the brain is more sensitive to early arrivals (typically direct sound and 1st reflections ? ) and the brain is very good at filtering out late arrivals (2nd, 3rd, 4th etc reflections)?

So after reading this I guessed that good diffusing can be highly effective in making 1st reflections become later arrivals ?

One key assumption I made was that good (broadband) diffusing is easier to achieve than good broadband absorption but I now see this can be a bad assumption.
First of all; the paper you attached contains several misunderstandings and use incorrect terms. It's too much to go in to at all, but you should take it with a grain of salt.

While it's true that the earliest reflections are most audible that doesn't mean that later ones aren't audible. They still are and the audibility of discrete refletions will highly depend on how much of the specular reflections are attenuated. If almost none are, you basically don't hear individually reflections but everything is chaotic and fuzzy. When you start treating specular reflections the remaining ones start to become audible (also depending on direction they come from). The paper below examines late arrival reflections from the rear wall in a LEDE room.
http://www.aes.org/e-lib/browse.cfm?elib=5242

Bottom line is that late arriving reflections are also audible and benefit from treatment. Generally I would recommend attenuating early arriving reflections and diffuse late arriving ones. The room dimension and size decides what's possible besides speaker directivity.

It's wortht noticing that in Blackbird Studio they achieved to highly attenuate even early reflections (most with 30 dB) with broadband diffusion because of the distance and volume of diffusers. The design is called ambechoic. It was really an experiment and aimed at primarily surround mixing. Most find the room quite dead compared to a modern LEDE or LEDE/RFZ design room.

ETC of Blackbird studio:
ETC.jpg
 

j_j

Addicted to Fun and Learning
Audio Luminary
Technical Expert
Joined
Oct 10, 2017
Messages
735
Likes
1,191
Location
My dining room.
#13
First of all; the paper you attached contains several misunderstandings and use incorrect terms. It's too much to go in to at all, but you should take it with a grain of salt.

While it's true that the earliest reflections are most audible that doesn't mean that later ones aren't audible. They still are and the audibility of discrete refletions will highly depend on how much of the specular reflections are attenuated. If almost none are, you basically don't hear individually reflections but everything is chaotic and fuzzy. When you start treating specular reflections the remaining ones start to become audible (also depending on direction they come from). The paper below examines late arrival reflections from the rear wall in a LEDE room.
http://www.aes.org/e-lib/browse.cfm?elib=5242

Bottom line is that late arriving reflections are also audible and benefit from treatment. Generally I would recommend attenuating early arriving reflections and diffuse late arriving ones. The room dimension and size decides what's possible besides speaker directivity.

It's wortht noticing that in Blackbird Studio they achieved to highly attenuate even early reflections (most with 30 dB) with broadband diffusion because of the distance and volume of diffusers. The design is called ambechoic. It was really an experiment and aimed at primarily surround mixing. Most find the room quite dead compared to a modern LEDE or LEDE/RFZ design room.

ETC of Blackbird studio:
View attachment 69162
Um, that paper does not say "first reflections are most audible" at all. So let's start there. It says "less audible", as in "mostly masked" except for frequency shaping effects. (and one of course must note that frequency shaping can affect localization, if I need to point that out?)

You note my reply to the OP did not say that first reflections are most audible, rather that they have rather less audble effect other than frequency shaping. The OP apparently got it backwards but my reply DID point out that the first reflections are mostly discounted. I don't think anyone is arguing that later reflections are inaudible, well, beyond OP's temporary misunderstanding. Are you denying precedence effect now?

And, yes, I wrote that slide deck.

I specifically deny that there are 'misunderstandings' in the slide deck, that it is 'misleading', and further that it asserts early reflections are more audible.

Finally, since the deck advocates treating the room first several times, especially if there are dominant modes, I'm not sure why you present the idea of treating later reflections as novel. The point of room correction, which the paper addresses, is to do what one can in a bad situation.

If one is actually designing a room, there is much more to consider.

Having scanned the Berger paper, I really don't see any disagreements here, either. The Berger paper appears to be mostly a (reasonable) criticism of LEDE, which has never been on my list of favorite things, just as an aside.
 
Last edited:

Bjorn

Senior Member
Joined
Feb 22, 2017
Messages
388
Likes
404
Location
Norway
#14
Um, that paper does not say "first reflections are most audible" at all. So let's start there. It says "less audible", as in "mostly masked" except for frequency shaping effects.

You note my reply to the OP did not say that first reflections are most audible, rather that they have rather less effect. The OP apparently got it backwards and I didn't notice, but my reply DID point out that the first reflections are mostly discounted. I don't think anyone is arguing that later reflections are inaudible, well, beyond OP's temporary misunderstanding.

And, yes, I wrote that slide deck.

I specifically deny that there are 'misunderstandings' in the slide deck, that it is 'misleading', and further that it asserts early reflections are more audible.

Finally, since the deck advocates treating the room first several times, especially if there are dominant modes, I'm not sure why you present the idea of treating later reflections as novel. The point of room correction, which the paper addresses, is to do what one can in a bad situation.

If one is actually designing a room, there is much more to consider.
You misunderstand. I wasn't referring to anything of the matter you specifically mention here when I said the paper contained misunderstanding and wrong terms. It's regarding other other areas and generally speaking.

Your latest paragraph (seen below too) also reveals a misunderstanding of fundamentals. But sorry, I dont' have time to go into the details.

Finally, since the deck advocates treating the room first several times, especially if there are dominant modes, I'm not sure why you present the idea of treating later reflections as novel. The point of room correction, which the paper addresses, is to do what one can in a bad situation.
 

j_j

Addicted to Fun and Learning
Audio Luminary
Technical Expert
Joined
Oct 10, 2017
Messages
735
Likes
1,191
Location
My dining room.
#15
You misunderstand. I wasn't referring to anything of the matter you specifically mention here when I said the paper contained misunderstanding and wrong terms. It's regarding other other areas and generally speaking.

Your latest paragraph (seen below too) also reveals a misunderstanding of fundamentals. But sorry, I dont' have time to go into the details.
So you are willing to make very serious professional accusations, but you are unwilling to specify why you made these accusations? Furthermore, "reveals a misunderstanding" which you are, strangely, unwilling to explain. Furthermore, when your initial misunderstanding of the slide deck in question is challenged, you claim "oh no, it's something else". But somehow, you can't seem to say what.

I think you've engaged in very irresponsible accusations here.

You are responsible for explaining these criticisms, and promptly.
 
Last edited:

Bjorn

Senior Member
Joined
Feb 22, 2017
Messages
388
Likes
404
Location
Norway
#16
Two concrete examples of misunderstandings in the paper is the information about "diffuse tail" and "RT60". The first isn't used correctly related to small room acoustics. The other isn't relevant to small acoustics at all. That's well known for experts in small room acoustics. But no, I don't have time to teach people online fundamentals which and takes a lot of time and bickering back and forth, and it would also be detrimental to this thread and the thread starter.
 
OP
M

Music1969

Major Contributor
Joined
Feb 19, 2018
Messages
1,043
Likes
325
Thread Starter #17
and it would also be detrimental to this thread and the thread starter.
It seems quite core to this thread because that paper is what started my thinking about diffusion (converting early reflections to later reflections), so I would have no issues if you went into more detail.

So if you wanted to expand, please feel free, when you have the time.
 

Bjorn

Senior Member
Joined
Feb 22, 2017
Messages
388
Likes
404
Location
Norway
#18
A "diffuse tail" is a term in small acoustics which is achieved with the treatment of diffusers in order to deal with spaced specular reflections.

The goal is to have a lateral late arriving diffuse tail which to a large degree emulates the best concert halls and yields spaciousness and envelopment without obscuring the recorded signal. You achieve both accuracy with a high degree of clarity, intelligibility, localization and correct tonality combined with a spacious and enveloping experience.

The ETC graph below illustrates it.
ETC after treatment with diffusion.jpg
 

j_j

Addicted to Fun and Learning
Audio Luminary
Technical Expert
Joined
Oct 10, 2017
Messages
735
Likes
1,191
Location
My dining room.
#19
Two concrete examples of misunderstandings in the paper is the information about "diffuse tail" and "RT60". The first isn't used correctly related to small room acoustics. The other isn't relevant to small acoustics at all. That's well known for experts in small room acoustics. But no, I don't have time to teach people online fundamentals which and takes a lot of time and bickering back and forth, and it would also be detrimental to this thread and the thread starter.
Wow, you really like to take things out of context. First, "small room" was not a given in the paper and your criticism is out of context. Then, of course, one can create a moderately diffuse tail in a smallish (not tiny) room with a lot of work (that isn't worth it for a couple of obvious reasons) but it's probably better not to have much of a tail (like I said, I'm not at all fond of LEDE), and finally, a T60 is absolutely measurable in such places unless they are too noisy, even though they are unlikely to be long, pleasant, or useful. Of course, I'd probably use T20 in most such places, but you already knew that, didn't you? Your claim to "doesn't apply" is simply evidence you don't know how to apply it as far as I'm concerned. I certainly labor under no such difficulties.

No, the OP's small room is not likely to be either diffuse initially nor have any useful T60 given the level of absorption he shows with his material. Does he want diffusion? What is his goal? We don't really know yet. That's a different issue to the perceptual issue that he asked about, now, isn't it? Yes, he had the perceptual issues wrong, that happens with people just starting out when they are learning. One of the ways people learn is by asking questions, right or wrong, and then a dialog an ensue, unless somebody drives by to pick a fight.

OP - what are you wanting to actually do in your room? That's where I was trying to get. Do you want imaging, envelopment, what? 2 channel, 5 channel, 7 channel, what? I was trying to work my way into finding out what you're thinking of when the bomb dropped. Do you know if you like LEDE-style for 2 channel?
A "diffuse tail" is a term in small acoustics which is achieved with the treatment of diffusers in order to deal with spaced specular reflections.
Yeah, I agree. It's what I showed the envelope of. Why do you feign disagreement?

So, then, your entire objection (which you could go to the lecture to listen to to find out) is that I drew the envelope of the diffuse tail? Any words about how diffusion happens, why it becomes probablistic, when it becomes probabilistic, etc, now? Would you be happier if I filled the envelope with white noise (I know, it wouldn't be white, but it would look pretty.).
 
Last edited:

j_j

Addicted to Fun and Learning
Audio Luminary
Technical Expert
Joined
Oct 10, 2017
Messages
735
Likes
1,191
Location
My dining room.
#20
It seems quite core to this thread because that paper is what started my thinking about diffusion (converting early reflections to later reflections), so I would have no issues if you went into more detail.

So if you wanted to expand, please feel free, when you have the time.
Well, he's welcome to continue to agree with me under the guise he is disagreeing :)

I'm still not sure exactly what you want out of the room? Is it stereo? Multichannel of some sort?
 
Top Bottom