If you look e.g. at the Shahinian "Hawk" module
- which is the top box for midrange and highs in the "Diaposon" model http://www.pearaudio.com/diapason.jpg -
then you can see Shahinian introducing some kind of phase decorrelation in highs at least:
http://www.fl-electronic.de/neuklang/images/shahawk.jpg
But this "upfiring" tweeter ensemble has higher directivity index than a single tweeter e.g. .
Nevertheless many people insist that Shahinian models like "Diaposon" and "Obelisk" were "omnidirectional", although these tweeter ensembles have much higher DI than conventional single tweeters of same size. The manufacturer in contrary names the speakers as "polydirectional".
If the tweeters were flashlights, one could agree they illuminate the room rather "omnidirectional". But tweeters aren't flashlights ...
Last edited:
IMO tweeters are like flashlights, only without the point source eg lamp. This difference is critical due to the wavelengths at hand and how it interacts with our dimensional constraints and design requirements. I would say that due to the physical limitations imposed, be it by inadvertant design flaw of having to have a large motor and surface area, mounting it to a baffle, even how it is loaded be it faceplate, short horn, waveguide etc all affect the radiation pattern eg DI. Is this any different than the reflector of a flashlight?
Seems if we put wavelength in perspective, a light emitting point source was from a source of equivalent wavelength to physical emitting size we would incure many of the same issues. If I'm not mistaken the band gap of an LED other than it's composition defines the wavelength emitted. In part this was how we squeeze blue light out of the little buggers.
Seems if we put wavelength in perspective, a light emitting point source was from a source of equivalent wavelength to physical emitting size we would incure many of the same issues. If I'm not mistaken the band gap of an LED other than it's composition defines the wavelength emitted. In part this was how we squeeze blue light out of the little buggers.
Light from usual (filament) lamps is uncorrelated. I cannot observe any interference in practical situations.
When mounting another tweeter on the same baffle more than e.g. a half wavelength apart from the first, there will be interference.
Even if the single tweeter radiates e.g. with DI = 3dB from a large baffle over the entire useable band of the tweeter (because the membrane is assumed very small compared to wavelength), the two imaginary tweeters at distance wont: You get a higher DI and a frequency dependent radiation pattern then, which is not "hemispherical" anymore.
This is why e.g. Shahinian "Hawk" module has higher DI than a single tweeter of the builtin ones would have.
The only chance of lowering DI is mounting tweeters "back to back" with distance small compared to wavelength. But any mounting at a considerable distance will increase DI even if you let the tweeters "point" in different directions. This holds at least as long as the membrane is small compared to wavelength:
Orientation of the tweeter is not relevant then, omitting "baffling effects" by larger faceplates etc.
When mounting another tweeter on the same baffle more than e.g. a half wavelength apart from the first, there will be interference.
Even if the single tweeter radiates e.g. with DI = 3dB from a large baffle over the entire useable band of the tweeter (because the membrane is assumed very small compared to wavelength), the two imaginary tweeters at distance wont: You get a higher DI and a frequency dependent radiation pattern then, which is not "hemispherical" anymore.
This is why e.g. Shahinian "Hawk" module has higher DI than a single tweeter of the builtin ones would have.
The only chance of lowering DI is mounting tweeters "back to back" with distance small compared to wavelength. But any mounting at a considerable distance will increase DI even if you let the tweeters "point" in different directions. This holds at least as long as the membrane is small compared to wavelength:
Orientation of the tweeter is not relevant then, omitting "baffling effects" by larger faceplates etc.
Last edited:
There are no discernable interference patterns only because the typical lamp is a broadband noise source where the radiating frequency is still ultra small compared to the source size. Tungstan lamps being warmer in color would be akin to pink noise than that of a coherent source like than of a laser and it's monochromatic emmision. Specturally balanced lamps like IR are akin to bass as UV is to a super tweeter.
Also to observe optical interference patterns you would need to dawn some polarized lenses unless you are related to the Mantas shrimp.
Also to observe optical interference patterns you would need to dawn some polarized lenses unless you are related to the Mantas shrimp.
Ok, I read through both papers and still I don't know what's the point, sorry
Can You please point me to the pages of the two papers relevant to the statement that "precedence does not depend on room reflections basic similarity* in time domain to the direct sound"?
(*because this is what I hypothesise about - not about them being identical)
To me it seems that such general statement is clearly beyond the scope of the studies described in those papers.
But I can be wrong of course and I like to learn. So please explain if I am wrong.
Last edited:
Wish we could use optical effects, especially Near IR. If this could be defined as just ordinary bass, then it would reradiate into the Far IR which would be infrabass, which is the subharmonic. But I'd be willing to bet if it were then the time lag and ringing would be far to great for our purposes. Could see military applications tho
/ot
/ot
I've read through this and I wonder what is exactly "good and poor acoustics" for Mr Griesinger?
Also I wonder if Mr Griesinger discredits all studies that lead to a rather established opinion that "Early reflections are known to be useful for speech intelligibility due to the fact that they can be partially integrated with the direct speech signal."?
The presentation reads a bit in such direction but I can be wrong of course. Please explain if I am wrong.
Very interesting but is this all really more relevant to the subject of HiFi than more focused research done for example by Toole and presented in His book or done by Bech in the framework of the Archimedes project?
Some unsorted thoughts ...
Concerning "off axis smoothness" of the frequency response of a loudspeaker:
What could be meant with "off axis smoothness" regarding the acoustic reality of a living room e.g. ?
No single reflections - e.g. from a booshelf placed at a side wall - will be "smooth" in spectrum. Even if the loudspeaker was radiating a "smooth spectrum" towards the shelf, in an appropriate angle to be reflected to the listening seat, that will not yield a reflection being "smooth" in spectrum: The spectrum of a single reflection will be "scattered" instead.
by "scattered" You really mean "filtered"? If not then what else?
Do You believe in a typical living room it would be "scattered" in the most important frequency range of
which I agree can be the
What about studies that showed that such systems were not preferred by listeners outside of the professional community which (as Floyd Toole puts it) can be actually regarded as professionally biased because of a certain unnatural oversensitization?
Last edited:
What You say reminds me of what Dave Moulton says:
OTOH His recommendations as to the loudspeaker design (which take into account the results of the Archimedes project) are quite different from Yours:
"
It seems to me that what we want to do is get all that information and then get no more information from the playback room. My design philosophy for studios is: let's have a perfectly reflective space for 50 milliseconds and then let's have no reflections or reverb after that. So let's have all the early delays with as little frequency response change and as little amplitude loss as possible, and then nothing after that."
If you want the early reflections and reverb to have similar spectrum - averaged over eligible directions of arrival and also time - like the direct sound but at the same time minimize (also destructive) interference, then making use of diffusivity seems a much better advice instead ...
That above citation is simply undefined in "how" that reflections are to be designed and thus is rather useless IMO.
I mean, what do you want: Closely related (or even "identical to direct sound" ?) reflections that arrive at the listener every 2ms e.g. at same level as the direct sound until 50ms elapsed ? Which are the prefered directions of the reflections ? Same as direct ? Maximum difference in angle ? Alternating ? 2 left 2 right ?
All from above ?
Most of those would be creating a horror scenario in every task that has to do with e.g. speech intellegibility and source separation.
That seems in line within the above citation, but frankly speaking such an approach is hardly to be taken serious. Also we have to distinguish, whether the room is used for performance ("production") or reproduction using e.g. a stereo setup.
In a reproduction scenario we might want to reproduce (or at least "make audible") the reverb of a recording venue, but we have to apply different criteria in the listening room itself ...
BTW: The above citation has no transparent relation to loudspeaker design at all.
It seems to me that what we want to do is get all that information and then get no more information from the playback room. My design philosophy for studios is: let's have a perfectly reflective space for 50 milliseconds and then let's have no reflections or reverb after that. So let's have all the early delays with as little frequency response change and as little amplitude loss as possible, and then nothing after that."
If you want the early reflections and reverb to have similar spectrum - averaged over eligible directions of arrival and also time - like the direct sound but at the same time minimize (also destructive) interference, then making use of diffusivity seems a much better advice instead ...
That above citation is simply undefined in "how" that reflections are to be designed and thus is rather useless IMO.
I mean, what do you want: Closely related (or even "identical to direct sound" ?) reflections that arrive at the listener every 2ms e.g. at same level as the direct sound until 50ms elapsed ? Which are the prefered directions of the reflections ? Same as direct ? Maximum difference in angle ? Alternating ? 2 left 2 right ?
All from above ?
Most of those would be creating a horror scenario in every task that has to do with e.g. speech intellegibility and source separation.
That seems in line within the above citation, but frankly speaking such an approach is hardly to be taken serious. Also we have to distinguish, whether the room is used for performance ("production") or reproduction using e.g. a stereo setup.
In a reproduction scenario we might want to reproduce (or at least "make audible") the reverb of a recording venue, but we have to apply different criteria in the listening room itself ...
BTW: The above citation has no transparent relation to loudspeaker design at all.
Last edited:
No problems with that, just not being aware Toole investigated cardiods explicitly ...
There is an example for rather mediocre if nod bad ranking of a certain ESL in Tooles studies AFAIK.
But "large area ESL" does not translate to "acoustic dipole" at mid and high frequencies, at least not in all cases. And one certain ESL under question had also weaknesses regarding one of the top criteria, which is "on axis smoothness" in response.
Anyway: Everytime a loudspeaker with "higher than usual" DI sounds "too dry" for part of the audience, changing speaker's and listener's position in a way to allow for lower D/R ratio is a way to go: Increase distance e.g.
As i pointed out, your own preferred DI from bass to midrange is about 6dB - when using edge position of loudspeakers in a room - which is pretty well in line with DI from cardioids e.g.
Your own listening habits thus seem closer related to "the dark side of the force" than you might assume ... maybe time to think about, why this is the case ?
Last edited:
It is clear in the context and it is defined verbatim elsewhere on Moulton's website:
Principles of Small Control-Room Layout
They should be designed as specular as possible:
*I believe that this reservation certainly holds true for any conventional forward firing speaker, but not for a FCUFS.
Well. It may seem so, but You might as well be surprised, just as Bob Olhsson was:
from: https://www.gearslutz.com/board/mastering-forum/139276-new-mastering-room.html
But isn't it that it is not that easy to achieve a sufficiently broadband diffusion? I mean effectively covering the important frequency range of 200Hz to 800Hz?
OTOH if it is not broadband enough then judged from perspective of an early reflection spectral content it just acts as a filter, not very unlike an absorber (see attached comment form Floyd Toole).
Attachments
Last edited:
It is directly related. Moulton's design concepts arose from experimental results of the Archimedes project.
I takes both room design and speaker design to achieve a specific early reflections pattern ...You know - loudspeakers and room as a system
(the original title of this thread)graaf said:
Diffusion seems only practical, say from midrange on upwards, that is true.
The more important is a sufficiently high DI from bass to midrange:
This is a displeasing requirement for loudspreaker designers, as it comes with higher effort due to construction.
Hard to succeed in a competition due to production cost, if you want to make progress in this respect.
But we are talking about requirements currently and not cost ...
Your own listening habits thus seem closer related to "the dark side of the force" than you might assume ... maybe time to think about, why this is the case ?
But I know why - I'm an "audiophile" after all ...for many many years now which means I am not a normal person anymore and I don't listen to music reproduction as a normal human anymore I am just as oversensitized as those professionals that Toole writes about.LineArray said:
...as a loudspeaker that has decorrelated radiation from upper midrange upwards to larger off axis angles will virtually "widen up" radiation to larger off axis angles (of course average DI will not get any lower by doing so).
In this way to the sides a high-pass effect with respect to reflected spectrum occurs which is contrary to the lowpass filter effect that would occur when placing (e.g. mid and high range) diffusers at the side walls and at the same time using conventional non directional loudspeakers from bass to midrange (which usually even tend to narrow radiation from upper mids to highs and thus further stimulate a "dull" quality of the reflected sound ...). *
In the proposed way of loudspeaker/room intercation, spectrally neutral and diffuse reflections - e.g. from the sides - are achieved, that cause minimum destructive interference with the loudspeakers direct sound.
___________
* In conventional 2-way loudspeakers often the range from 3 to 5 Khz is overly pronounced in the reflections, due to energy response rising at crossover frequency, while having a depression at upper mids/ lower highs.
This is part of the "most typical" loudspeaker/room interaction, which is an artifact observed in many living rooms.
Last edited:
I get Your point, ok. Still I am not convinced that loudspeaker design - as far as spatial aspects of reproduction are concerned - should fight the evil that the listening room does, the evil that lateral specular reflection do, and that it should aim at taking the room out of the sound equation with the use of higher DI, diffusive wall treatments and so on.
You mean DML or similar reproducers? Ok, I get Your point. Still I don't know what is "destructive" about coherent lateral reflections, ipsilateral in particular. Studies by Toole, Bech and others suggest rather that there is nothing "destructive" about them.
Last edited:
graaf said:
... well, how to take out of the equation "the evil that the room does".
Who should do it then, if not the loudspeaker ?
The readiness of listeners for effective room treatment in the home is very low in fact. Mostly you have a living room as the listening room and you also have to take it - more or less - "as it is".
You can promote some left to right symmetry maybe in speaker placement, and if e.g. reflective and diffusive conditions at the side walls are fairly symmetric, that's a bonus: Fine.
But in most cases it is not. Often also the side reflections are more early than they should be: Sometimes you simply cannot change that, because of other restrictions arising from the fact, that the room serves as a living room in first place.
A loudspeaker sending the message:
"Hey, i am a diva, please adjust delay of reflections and install some diffusers. Furthermore please correct for my uneven energy response ..."
Is not good to handle IMO.
A loudspeaker should be intentionally constructed to cope with typical room and setup issues, namely rather short distances to the walls, lack of diffusivity etc.
If the room is better than expected or the owner is willing to do some treatment, then the speaker should be able giving back a considerably nice reward ... a "robustly" designed loudspeaker will do that even to larger extent than a "diva".
Last edited: