Diffussion and dipole speakers

[boggy]

Originally Posted by HIFI1959

I am asking about my front wall that my ESS AMT Dipole speakers are placed. My AMT Dipole driver is crossed and operating 800 Hz and up. The driver is 43 inches away from front wall. Is there any diffusion or absorb I should consider?

In the small rooms absorption is needed below roughly 500Hz, because resonant and non resonant interferences caused by room boundaries. From my experience, for dipole speakers and their back fired wave, diffusion above 500Hz is better than plain wall reflection or absorption.

Originally Posted by HIFI1959

Will diffusion work at all as close as driver is to wall?

Yes, at least it will work better than wall only.

[diffusor.com]

Broadbanded temporal diffusion is with my experients , the key in small room acoustics
I agree with Boggys advice, but only if you use ordinary diffusors
My toolbox is very different becouse the concept consist of halfwave delay lines that create a broadbanded diffuse field just infront of them
You can place them as close as 12" from the speaker
This modules cant be calculated with BEM, instead we had to calculate them in CATT
A CATT example , S-Wing module delay sound as as low as 250Hz

left flat wall right S-Wing module

http://i386.photobucket.com/albums/o...gcatt250hz.jpg

3 S-wing evenout the frequence by temporal diffusion Mike 12 " from the wall
red curve flat wall
Green curve 3 Wing modules side by side

http://i386.photobucket.com/albums/o...reochefter.jpg

Best
Matts Odemalm

SMT AB

[boggy]

Originally Posted by diffusor.com

......
I agree with Boggys advice, but only if you use ordinary diffusors ......

Yes, I mean "ordinary" phase/amplitude grating diffusers or any other construction derived from QRD (PRD) and binary diffusers, 1D or 2D.

[Jens Eklund]

Originally Posted by diffusor.com

This modules cant be calculated with BEM, instead we had to calculate them in CATT

As far as I know, CATT (or any other room acoustic simulation software that I know of) does not simulate diffraction in such a way that you can predict diffuser performance (you can only simulate the effects of diffusing surfaces in general assuming the scattering coefficients entered are correct *). The only (commercially available) software that I know of that can simulate diffraction in such detail so that you can use it to simulate diffuser performance properly, is "Reflex" from AFMG (since it uses BEM models and can therefore properly simulate diffraction in detail). Please note that this is not meant to critzise CATT or other room acoustic simulation software packages, if you were to use BEM (or FEM) for a complete room, it would take ages to compute, especially for higher frequencies so this is not very realistic to expect. Refelx does not do room simulation, but focuses on precise simulation of scattering surfaces (including diffraction effects which is crucial for a scattering devise) and even if very limited geometry, the simulation can still take hours to compute even on a modern multicore machine.


Comment on CATT when trying to simulate a diffuser in detail:

"- Modeling the diffuse surface exactly failed to reproduce the measured response. This is important. It is not a shortcoming of the modeling program, but a shortcoming of modeling sound as a ray and not including the wave nature of sound."

Room Modeling Details



* CATT currently also supports 1D-scattering in the prediction model:
http://www.catt.se/Lambert-1D-CATT.pdf

[boggy]

Originally Posted by Jens Eklund

......The only (commercially available) software that I know of that can simulate diffraction in such detail so that you can use it to simulate diffuser performance properly, is "Reflex" from AFMG.
.......

I'm curious, is it possible to define absorption coefficient of each simulated diffuser segment in Reflex? Sorry for OFF-TOPIC.

[Jens Eklund]

Originally Posted by boggy

I'm curious, it is possible to define absorption coefficient of each simulated diffuser segment in Reflex? Sorry for OFF-TOPIC.

At the moment; no.

[Jeffrey Hedback]

Odeon seems to be able to model more complex in-room diffusor applications/responses...if someone can afford it and wants to pursue, I'd love to explore.

[Jens Eklund]

Originally Posted by Jeffrey Hedback

Odeon seems to be able to model more complex in-room diffusor applications/responses...if someone can afford it and wants to pursue, I'd love to explore.

Without knowing for sure, I would be surprised if it´s not very similar to CATT and other programs. Being able to include diffusive elements does not mean that they are accurately modeled including precise diffraction effects. That would simply take too much processing power to compute. Perhaps in couple of years …

[Jeffrey Hedback]

Acoustics in Music studio | Odeon

It seems to be step forward from CATT. But the price of admission is a deal-breaker into seeing where the advances may exist.

We are on the cusp of even greater modeling power...agreed Jens!

[Jens Eklund]

Originally Posted by Jeffrey Hedback

Acoustics in Music studio | Odeon

It seems to be step forward from CATT. But the price of admission is a deal-breaker into seeing where the advances may exist.

We are on the cusp of even greater modeling power...agreed Jens!

Without being an expert on the algorithms used in these programs, I don´t see a big difference between CATT and ODEON:
http://www.odeon.dk/pdf/CLC%20fa2005.pdf

Neither of them uses full BEM/FEM (or possibly FDTD, Finite Difference Time Domain) to properly model diffraction which makes sense since the models would then become extremely complex and require a tremendous amount of computing power.

But even with a perfect model able to simulate everything, and infinite processing power; the model relies on the input data to be accurate in order to produce accurate predictions, and the more complicated a model becomes, the more input data is required and a lot of data is hard to source in such detail, like the exact complex impedance of a boundary for instance.

Buy yes of course; a lot can improve and this is just the beginning ...


Sincerely Jens Eklund

[HIFI1959]

Thanks for all help and thanks in advance if you continue.

So what next . . .

I cant audition Diffusion and this stuff is not cheep. What brand product would one recommend? What about placement on my 11.5 ft wide wall?

I have eye on ASC TubeTraps and that company recommended one each 16" X 8' for each front corner and one each 13" X 7' to be placed behind speakers on front wall. Both Bass Traps have diffussers built in. I had a second company recomend two absorer and to diffusor for front wall. Each piece was 2ft X 4ft. Both companies have my room specs.

Am I lost thinking I can make my stage image appear larger than my 11.5 ft boundry? If I am it was still worth a try.

[diffusor.com]

Jens you have to let go how ordinary mathematical broadbanded diffusors are evaluated
To make a long story short , my experients are if you want to expose everything in a recording and be transported to the recording room , a broadbanded S-field must be created by modules that not need a neigbour to do the job
Thats not possible with "ordinary" diffusion modules becouse they need a certain listening distance becouse of the deep wells and lots of friends todo the job
Instead I begun to expriment with different long pathways (open in both ends ) and thats why BEM cant be used
The broadbanded halfwave delayline family (Wing Family) was actually ready when Welcome to Audio Data Lab run a module in CATT , just to see how long and even different frequences are delayed inside the module So instead of searching for the "perfect" mathematical pattern (that humans probobly never will get used to if its not fractal ) I tryed to activate the brain with reflections as much as possible using different lenghts on the delay lines
ARTA is highly recommended and the measurments I looked for in ARTA during evaluation of the Wing family, was spot on the results from CATT
One of many catt calculations shows one singel S-wing (600*1200*180mm w*h*d) 4000hz
flat wall left S-wing right


Accuracy needs a room in balance and that means that the "room frequence respons " should be a copy of the direct sound Its easy to trunk the room respons in ARTA but its very rare this very important respons is documented What I have seen on this forum are McTwins the only ones that show the frequence respons from the room

Best
Matts
SMT AB

[Jens Eklund]

Originally Posted by diffusor.com

Jens you have to let go how ordinary mathematical broadbanded diffusors are evaluated
To make a long story short , my experients are if you want to expose everything in a recording and be transported to the recording room , a broadbanded S-field must be created by modules that not need a neigbour to do the job
Thats not possible with "ordinary" diffusion modules becouse they need a certain listening distance becouse of the deep wells and lots of friends todo the job
Instead I begun to expriment with different long pathways (open in both ends ) and thats why BEM cant be used
The broadbanded halfwave delayline family (Wing Family) was actually ready when Welcome to Audio Data Lab run a module in CATT , just to see how long and even different frequences are delayed inside the module So instead of searching for the "perfect" mathematical pattern (that humans probobly never will get used to if its not fractal ) I tryed to activate the brain with reflections as much as possible using different lenghts on the delay lines
ARTA is highly recommended and the measurments I looked for in ARTA during evaluation of the Wing family, was spot on the results from CATT
One of many catt calculations shows one singel S-wing (600*1200*180mm w*h*d) 4000hz
flat wall left S-wing right


Accuracy needs a room in balance and that means that the "room frequence respons " should be a copy of the direct sound Its easy to trunk the room respons in ARTA but its very rare this very important respons is documented What I have seen on this forum are McTwins the only ones that show the frequence respons from the room

Best
Matts

Matts,


What I´m trying to say in post 5 is that you cannot (accurately) model a diffuser in CATT (well, you can, but the result will not reflect real would performance since you need an accurate diffraction model in order to simulate scattering surfaces). The synaudcon-article explains why.

I´m struggling to follow your logic regarding your conclusions, but if you feel that your illustrations can prove something, then please repost them in a higher resolution so one can actually see what they are showing (I can see that it´s the simulated echogram from CATT (Schröder (abs.)), but what are you suggesting it proves?) and if you could explain why you think CATT can simulate your diffuser accurately, then please do. I´ve recently had a number of conversations with Bengt-Inge (the creator of CATT) and I would love to hear what he has to say about it all (although I think he shares my view on the matter).

And let’s not discuss different acoustic design concepts, I know that you push your own (in my opinion somewhat controversial model if studio control rooms), and we would never agree on anything so let’s not even start . Let´s just focus on diffuser performance prediction alternatives. If needed, I could also ask for Peter D'Antonios opinion, but he normally stays far away from forums … (I ´m starting to understand why).


Sincerely Jens Eklund

[HIFI1959]

I have heard of folks using Artificial Ficus trees behind and to the outside of dipoles. Is this myth?

[Ethan Winer]

Originally Posted by HIFI1959

I have heard of folks using Artificial Ficus trees behind and to the outside of dipoles. Is this myth?

It's not a myth that people do this, but it's a myth that a tree is useful as acoustic treatment.

--Ethan

The Acoustic Treatment Experts

[George Necola]

Originally Posted by Ethan Winer

It's not a myth that people do this, but it's a myth that a tree is useful as acoustic treatment.

--Ethan

The Acoustic Treatment Experts

Why? I recorded black metal drums in a forrest (no joke), and the trees were really diffusing the sound.



Gesendet von meinem GT-I9100 mit Tapatalk

[boggy]

Originally Posted by HIFI1959

I have heard of folks using Artificial Ficus trees behind and to the outside of dipoles. Is this myth?

Diffusers are deterministic "creatures"... they must work in the best possible fashion, in same way, every day in week, regardles of the season, regardless you care enough for it and regardless of temperature, lighting, and humidity in the room.

So trees/plants are ok, but we still needs diffusers instead of only plants.