I don't think that's necessarily that difficult.
This is with a speaker with a 12" horn and woofer crossed at 1 khz in a relatively live room of about 100 m^3, with damping panels at the first reflection points.
Attachments
Yep, me too. It depends on the hall, of course, but even in very live spaces precision is still possible. I have a good example involving a flute, a parasol and the Maui High School auditorium, but won't bore you with that long story.
I disagree. In concert halls I've never had the same perception of pin point imaging I get with my Nathans. The feeling of being able to localize instruments is probably just another example how easily our hearing is overpowered by our sense of sight.
Geddes Nathan without absorber at side walls:
An externally hosted image should be here but it was not working when we last tested it.
People do not seem to understand the deffinition of "difuse". An impulse response will not show if the sound is difuse or not. Diffuse means that there is no dominate direction to the sound. A simple impulse response cannot tell you that. "Difuse" implies that if the intensity (a vector) at a point is integrated over direction it will be zero. In other words the time averaged intensity at any point is zero. Its actually not a criteria for listening at all, I am not sure why it is even discussed in that context. It is an assumption made (and required) for reverberant room testing of materials since this is the only way that the math can be worked out.
People do understand the term diffuse (homogeneous and isotropic) but the ETC posts were about your comment that it's "difficult to impossible in most small rooms" to achieve 15dB/15ms.
What I personally don't like is the sharp window you are hearing through into the auditory scene. But maybe the typical small room is just too small for such speakers and people sit too close. Things might change for the better if you would sit farther away.
My trial of an explanation.
I was more reacting to the often quoted oppinion that domestic rooms are never proper reverberent fields (hard not to use the term diffuse). They clearly would be above the Schroeder frequency.
I'm not sure what the most rigourous definition of diffuse is, but it is usually used in the context of surface diffusion and the property of breaking up a wave and reducing specular reflections. Doesn't it also mean that, after a few reflections, sound will come from all directions with no strong non-random components (such as slap echo or unidirectional modes)?
David
That's even markedly cleaner than I get when I measure speakers outside. Before I had absorbers, there were one or two reflections louder than -15 dB, the contralateral one being the most pronounced.
EDIT: Or maybe I don't know how to read your plot. The impulse peak is probably not at zero dB's. Maybe at -30 dB and about 114 ms?
Last edited:
I was refering to the fact that getting no reflections < 15 ms. is difficult in a small room (that's not anechoic). I don't think that either set of data shows that. Both are too coarse at small ms to really see for sure.
Last edited:
Toole: "Classic concert-hall acoustical theory begins with the
simplifying assumption that the sound field throughout a
large relatively reverberant space is diffuse. In technical
terms that means it is homogeneous (the same everywhere
in the space) and isotropic (with sound energy arriving at
every point equally from all directions)."
Concert halls aren't 100% diffuse, acoustically small rooms even less so. Diffusivity isn't a helpful measure of what is perceived.
Last edited:
I guess that I don't know what this means either, so it's not much of an explaination to me.
No reflections <15ms isn't going to happen in a free field either. There's always some reflection from the floor/ground. The working hypothesis is that we don't need to eliminate reflections completely (to remove them perceptually) but just need to lower them beyond threshold, which is more like -25dB. I would agree that this number is really hard to achieve.
Here's a zoomed in ETC of my room (left speaker at main listening position):
An externally hosted image should be here but it was not working when we last tested it.
Last edited:
This is how I listen to the largely omni Pluto 2.1 speaker in a small room:
Linkwitz-Links
"Loudspeaker Placement for Optimized Phantom Source Reproduction
A joint paper by J. Gerhard, B. Theiss, M. O. Hawksford from the 1996 AES Convention in Copenhagen (Preprint 4246). Their loudspeaker layout is quite different from the ORION++, but similarly based on psychoacoustic observations. I suspect that their listening experience could have been even more convincing, if omni-directional loudspeakers had been used. Those would have increased the reflected energy from the wall in front of the listener over the energy coming from the side walls. That would have reduced the need for micro placement of speakers and listener to 0.5" (12 mm) accuracy without affecting imaging precision.
A setup of loudspeakers parallel to the long wall of the room and close to the listener would seem to be well suited for small rooms and the PLUTO. It could also easily fulfill the minimum distance requirement of 3' (6 ms) from any wall. Close listening distance to the speakers also reduces the magnitude of floor and ceiling reflections if that should matter. "
Markus - I misread the scale - yes that is remarkably good in-room results.
I don't believe Toole is claiming concert halls are diffuse so much as he is pointing out that the Sabine notions must assume that they are.
A great deal of concert hall design these days has to do with the subtle ways in which real performance spaces deviate from the ideal. For example an opera house (and to a lesser extent a concert hall) is best considered as 3 linked spaces, the orchestra pit, the proscenium and the main floor. Modeling it as a large volume with the bulk absorption properties of all surfaces will be kinda close, but not always good enough to guarantee a first rate hall. The modern ray tracing programs for acoustics get around these issues and come up with a more accurate result when classical acoustics isn't close enough.
Beranek quotes diffusion as being a significant variable and shows correlation between high and low diffusion in halls and their perceived quality. It isn't the biggest variable but it is on the list.
That we discuss concert halls as diffuse relative to domestic rooms is because the Schroeder frequency is so low in a large room, while it falls a 1/3rd of the way up the frequency range in our living room. True surface diffusivity should be higher in a concert hall than at home, simply because a concert hall needs to sound good and they put the effort into making it so. (Beranek still attributes a lack of diffusion as part of the Philharmonic Hall disaster.)
I will say that I've measure RT in my living room several times and several ways and I always get consistent numbers above 160 Hz for a given method, nomatter where the source and mic are. I would think that implies a diffuse field.
David S.
Hi Guys
I was thinking about that left minus right test some of you had done, that had my curiosity going. I had an Arta set up in my livingroom the other day and thought hey, I will do that test. I figure these may look different as a result of having a little more directivity than most hifi speakers.
The speakers are my old passive xover SH-50’s, built about 7 years ago and could probably be refreshed. They are a three way full range horn employing 7 Drivers. The room is fairly cluttered with mine and my kids stuff and there is no room treatment at all with a bare hardwood floor.
There is a wall about 2 feet behind the speaker and a secretary desk and the rear of a motorcycle within a foot of the mouth (IE; not a pristine anechoic condition).
The directivity is such that there should not be any significant short side reflections, while the pattern is 50 degrees wide, it is still -20dB at 90 degrees off axis at 500Hz and about -10dB 90 degrees off at 300Hz and they are aimed inward relative to the side walls.
The listening position is a couch; I have the microphone stand at the foot of the couch, center seat but a bit higher than seated height (wanting to avoid couch effects). The rear wall is well behind the couch.
Picture one is the impulse response for both SH-50’s in polarity and both SH-50’s out.
The listening position is about 14 feet from the speaker, the mic is 18.8ms away from the origin of sound within the speaker. The yellow curve is the impulse response for the two speakers in polarity and the green is the out of polarity case.
As one can see the speakers cancel each other out well at least initially but after a very short time, the stray sound begins.
Picture two is the magnitude responses at LP both in and out of polarity condition.
Being an old TEF guy, I like the ETC view of time better than the impulse (the real component of the envelope), the ETC shows energy vs time.
The next picture is the ETC from the right channel at the LP.
Notice the obvious first arrival and a couple ms later, the floor bounce arrival.
In placing the speakers “back then” I put them on top of the subwoofers which at mid height, made the ceiling bounce and floor bounce add together where the mic is at mid height, clever planning eh…ugh?
A closer distance or narrower vertical pattern would be better.
Anyway some curves to add to the heap.
Best,
Tom Danley
Danley Sound Labs
Danley Sound Labs, Inc. | Facebook
I was thinking about that left minus right test some of you had done, that had my curiosity going. I had an Arta set up in my livingroom the other day and thought hey, I will do that test. I figure these may look different as a result of having a little more directivity than most hifi speakers.
The speakers are my old passive xover SH-50’s, built about 7 years ago and could probably be refreshed. They are a three way full range horn employing 7 Drivers. The room is fairly cluttered with mine and my kids stuff and there is no room treatment at all with a bare hardwood floor.
There is a wall about 2 feet behind the speaker and a secretary desk and the rear of a motorcycle within a foot of the mouth (IE; not a pristine anechoic condition).
The directivity is such that there should not be any significant short side reflections, while the pattern is 50 degrees wide, it is still -20dB at 90 degrees off axis at 500Hz and about -10dB 90 degrees off at 300Hz and they are aimed inward relative to the side walls.
The listening position is a couch; I have the microphone stand at the foot of the couch, center seat but a bit higher than seated height (wanting to avoid couch effects). The rear wall is well behind the couch.
Picture one is the impulse response for both SH-50’s in polarity and both SH-50’s out.
The listening position is about 14 feet from the speaker, the mic is 18.8ms away from the origin of sound within the speaker. The yellow curve is the impulse response for the two speakers in polarity and the green is the out of polarity case.
As one can see the speakers cancel each other out well at least initially but after a very short time, the stray sound begins.
Picture two is the magnitude responses at LP both in and out of polarity condition.
Being an old TEF guy, I like the ETC view of time better than the impulse (the real component of the envelope), the ETC shows energy vs time.
The next picture is the ETC from the right channel at the LP.
Notice the obvious first arrival and a couple ms later, the floor bounce arrival.
In placing the speakers “back then” I put them on top of the subwoofers which at mid height, made the ceiling bounce and floor bounce add together where the mic is at mid height, clever planning eh…ugh?
A closer distance or narrower vertical pattern would be better.
Anyway some curves to add to the heap.
Best,
Tom Danley
Danley Sound Labs
Danley Sound Labs, Inc. | Facebook
Attachments
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.