Acoustics when speaker dispersion pattern is known

If room acoustics below 800 Hz were not an issue then ignoring what happens at those frequencies might make some sense. In fact substandard room acoustics below 800 Hz can be a big issues.

You'd probably like to be a fly on the wall when I discuss this issue with a certain well-known designer and manufacturer of controlled-directivity speakers. IMO directionality control above 800 Hz is a great idea, such a good idea that ideally it would be continued down to 80 Hz.

Besides, you get my second Pointed Cap Award of the day for over thinking the problem. What do your acoustical measurements of your room tell you? Contrary to what seems to be a popular belief, audio isn't just about head games. Getting your hands dirty with a little acoustical test equipment could change your life, and I mean that in a good way.

I wish I knew enough to be able to plan acoustic treatments without measuring. I don't know what specifically should be considered, but I can point you to a couple important facts and considerations.

First, I wouldn't expect to leave the speakers "square" in the room. Some (maybe a lot) of toe-in is likely best. That won't change the positions of the reflection points, but it will change the magnitude and frequency profile of those reflections - shifting them toward on-axis for some and away for others.

Second, the transition to modal response (away from specular response) will occur near the Schroeder frequency, which is different for each space. Commonly, the transition region (not a particular frequency - as the behavior changes over a range) in normal sized home theater spaces and listening rooms is between approximately 200Hz and 400Hz - larger rooms have lower ranges and smaller rooms have higher ranges. Your room would have to be exceedingly small to transition at 800Hz, so treating the off-axis sound is still part of the consideration. Remember that 200-800Hz covers a lot of critical frequency ranges, including a lot of musical tones and voices.

Third, since you have a little time before any of this needs to be finalized, I'd suggest you read Floyd Toole's book - Sound Reproduction: The Acoustics and Psychoacoustics of Loudspeakers and Rooms. At $50US (or so) it's a great investment (but I was able to borrow a copy from a local university library). He provides some sample layouts of treatments, based on patterns of directivity. His requirements for directivity are very stringent - He wants, especially for surround usage, speakers with exceptionally wide dispersion. Some of his sample designs call for speakers with uniform dispersion (as uniform as possible) over nearly 180 horizontal degrees. You will also find, reading his book, that there is a lot of evidence to suggest that treating first reflection points may not be your best choice.

Fred

I probably wouldn't worry about the "opposite speaker" reflection points, but the "same side" reflection points should be addressed. If you want to test this before going to the trouble and expense to buy or make absorbers, hang folded-over bath towels at the same side reflection points and see if that helps. I imagine it will. Then you can replace those with better absorbers. BTW, these reflections are of concern down to about 300 Hz.

--Ethan

There would be no reason to treat that reflection point even if your speaker didn't have controlled directivity. The reflections from the side walls in listening tests is shown to be a pleasing effect, widening the image beyond the specific location of the speaker. Here is an article I wrote on this topic with research references to back this: http://www.madronadigital.com/Library/RoomReflections.html.

And while you should measure your room to correct for bass frequencies, the need to do so for frequencies above transition is far lower. If there are problems there, they are problems of speakers (unless you have put way too much absorption in your room). On use of ETC, I address that in the article also. As you note there, that is not a fruitful path either.

The strategies here need to be based on perceptual effects and how we hear sound in rooms as opposed to pleasing graphs and meters for the sake of doing so . You can actually degrade the sound in your room as you make the graphs prettier!

I see a "thumbs up" icon to Like a post, but why is there no "thumbs down" icon to identify a troll post?

--Ethan

A lot of things can degrade the sound in a room. Making measurements is not one of them. Measurements provide information. Information is a good thing. The difficulty, as with anything else in life, is in the understanding, interpretation, and application of information.

Nothing is degraded until you take action. If you don't understand the intracacies of room acoustics (and no one expects you to), ask lots of questions here or elsewhere or hire an expert to help make decisions prior to taking action. I guarantee, the more information you can provide, the more fruitful the advice you will receive.

Don't let certain of our scaremongering members here rob you of the joy and enrichment that comes from expanding your understanding of something you are interested in. If audio is a hobby, the journey can be as enjoyable as the destination. Taking measurements along the way is essential to helping you grow that understanding.

For presumably cross-fired speakers that have fairly narrow directivity?

I'm on the side that says measure, then if you have to (after aiming the speakers, calibrating multiple subwoofers, etc.) move on to room mutilation only if you must.

I don't know what cross-fired means. Front loudspeakers should be angled toward the listener. Sound that goes past the listener will eventually strike the rear wall, which is another important place for absorption (or diffusion). I suppose in a very long room the direct sound might hit the rear of a side wall. But then the next bounce hits the rear wall.

I don't understand that either.

--Ethan

No, they shouldn't, if one wants a wider listening area and has speakers with well-enough controlled directivity to take advantage of time-intensity trading.. See Bill Waslo's excellent article, which I've reposted here.

The bottom line is a narrow-directivity speaker (and 90deg is fairly narrow) that's crossfired simply won't illuminate the same-side wall enough to matter, because the same-side wall will be far out of the pattern in the relevant frequencies.

Unless one prefers the spaciousness that comes from natural reflections, of course.

Acoustic products tend to be ugly, so I call using them "room mutilation."

I understand what he's aiming for, but I disagree with over-toeing speakers because that exaggerates loudspeaker beaming and lobing. I'm mainly concerned with getting a flat response that doesn't change drastically with small amounts of head movement.

The best way to achieve a wide image - not unnaturally wide, only as wide as the producer and engineer intended - is to absorb all side-wall reflections. Otherwise, the early "small room" ambience drowns out the larger, wider ambience that's already present in the recording. If one isn't allowed to put acoustic treatment in the living room there are ways to hide it, though of course that costs more.

--Ethan

No, it doesn't, assuming the speakers were competently designed in the first place. (And if there's a directivity shift in the midband, the problem is simply incompetent loudspeaker design.)

Using appropriate speakers set up using the Geddes/Waslo crossfiring approach (and multisubs), one can walk 5 paces in any direction without worrying about changes.

(Though in the same breath, the only speakers I've ever heard that really did change dramatically with small head movements were the big old Dunlavies.)

A hard center also helps immensely in that regard. (I never listen at home in stereo, it just sounds so...lesser than even synthesized - DPL2 - multichannel.)

That's a great way to suck the life out of a room, but isn't much good for anything else.

Our point of agreement here, I think, is that adding room mutilation products will actually have a sonic effect, unless say swapping amps (assuming both are adequate to the job) or wires, etc.

That "allowed" bit assumes men don't care how their homes look. I think that's a stupid assumption.

While I understand your current situation only allows you to explore the theoretical, there is no substitute for listening, especially when it comes to answering your question. Despite what the manufacturer of your speakers might say about dispersion, the only way to know whether you prefer their contralateral reflections absorbed is to try it both ways.

Personally, I've found that I prefer the ipsilateral reflections left alone (or maybe a bit o' diffusion) in order to broaden the soundstage. It's less obvious with the contralateral reflections but, given the choice, I prefer them absorbed (I don't like it when sounds coming from one side of the soundstage are heard, however subtly, from the opposite side as well).

But I would never have known that without listening. No amount of theorizing about it would have informed my preference. Once you get your system (or even just a couple of speakers) up and running, dial in the toe-in that you like, and then try absorbing the contralateral reflections to see if you prefer it.

I look forward to your posting some polar plots as evidence.

In the mean time, read and learn:

http://www.linkwitzlab.com/Constant_directivity_louds.htm

--Ethan

Evidence that speakers with midband directivity discontinuities (i.e. the typical "mushroom cloud midrange" of a speaker with a 6"+ midwoofer and a flush-mounted tweeter, with a typical mid-tweet crossover) are a priori incompetent designs, that may benefit more from the band-aids offered by room-mutilation products than competently-designed speakers do?

There are plenty of polar plots of competently-designed speakers about. But to hear the difference one needs to listen.

And if you're going to cite Linkwitz, how 'bout you post a picture of his room, showing the copious quantities of room mutilation products used in it.
Edited by DS-21 - 11/21/12 at 3:11pm