The myth of generic optimum room dimension ratios

[j_j]

The ultimate solution does exist which is to deal with all the laws of physics and fluids and fully model every aspect of the room and the speakers that excite it.

And, which ultimately results in anechoic listening rooms with electronic systems designed to simulate a given type of room.

Fortunately (now we can move into perception a minute) our ears and brains are quite good at separating out some playback room effects, thanks to a variety of mechanisms from short-term loudness adaptation in the periphery to various cognitive filtering techniques.

 

[j_j]

What I mean is that paper and theory is one thing, reality is another. People who look at the paper and at measurements often forget that a room built from real materials probably does not behave like the one on the paper drawing,

Yeah, that's a big mouthful, and it's rather an important point. Rather than go into detail, I'll just say that if you have something designed properly for acoustic properties and isolation, and you're having it built, watch like a hawk. Proper acoustic construction is not what most construction crews expect to do, and they will often do what they think is "right" instead, in ways that are very hard to remediate without ripping it all out.* Furthermore, real materials are not infinitely rigid and massive like most room design algorithms treat them.

and they often don't consider the psychacoustics.

Klaus

More like "almost always don't consider the psychoacoustic issues" or even a considerably number of the real acoustic issues. Not that I'm disagreeing here...


*When building one listening room that was double-wall for isolation, I was treated to the opinions of the carpenter who had to crawl across the interior ceiling and cut out all of the "braces" he added between inner and outer walls. The invective was quite impressive, as was his devout belief that having the walls uncoupled was the wrong answer. None the less, we got the room to between NC8 and NC9 in a rather noisy building. (floating floor and the works thrown at it) Using a wedge, it was easy to show that a wedge that coupled the floor at one point pushed us up to near NC20, all of it low-frequency noise. Had we not caught the "improvement" to the design before the final drywall was up ...

 

[amirm]

Fortunately (now we can move into perception a minute) our ears and brains are quite good at separating out some playback room effects, thanks to a variety of mechanisms from short-term loudness adaptation in the periphery to various cognitive filtering techniques.

Indeed. Our perception of room response often is anything but what we expect above the transition frequencies of a few hundred hertz. It is challenging to get people to not follow simple measurements of reflections over these important psychoacoustics.

 

[j_j]

Indeed. Our perception of room response often is anything but what we expect above the transition frequencies of a few hundred hertz. It is challenging to get people to not follow simple measurements of reflections over these important psychoacoustics.

It is possible to make relevant measurements above the wavelength to precedence transition, but that's not what most (?any?) systems do.

 

[Steve Williams]

Proper acoustic construction is not what most construction crews expect to do, and they will often do what they think is "right" instead, in ways that are very hard to remediate without ripping it all out.* Furthermore, real materials are not infinitely rigid and massive like most room design algorithms treat them.

Having just finished the build out of my new room, I agree completely with this

Me and my acoustician had to meet with the contractor and all of the sub contractors twice before the build out began to be completely certain that they knew exactly how we wanted things built

 

[microstrip]

There are two schools here:

1. This class says there are magic ratios like golden dimensions and tell you to chase them. So if you believe such a rule must exist, then people are providing the specifics.

2. This class says the analysis that group #1 performed is too simplistic and cannot be relied upon to matter in real life (e.g. you don't sit in a corner of a room and put the speaker in the other corner). The method that is proposed is to let the room be whatever it is, then come up with systems the give us the results we want through placement, use of subs and number/placement of those, plus electronic correction (and I don't just mean EQ). The fact that we give up on formulas and criteria for good/bad rooms is not an admission of defeat but rather focusing on solutions that really work, as opposed to simplistic formulas that don't.

The ultimate solution does exist which is to deal with all the laws of physics and fluids and fully model every aspect of the room and the speakers that excite it. Such complete modelling is too expensive and labor intensive to be practical for every room of even most high-end rooms. So we take short cuts and other tools per above.

Amir,
Your answer just disguises that we are not able to answer a direct question that is often asked - I am building a room from scratch mainly for stereo listening. What dimensions should it have in order to get optimum results with minimal problems to be solved using multi-subs and equalization? Such that I will not need to spend usd 10K to simulate my room later in order to optimize it?

 

[j_j]

Amir,
Your answer just disguises that we are not able to answer a direct question that is often asked - I am building a room from scratch mainly for stereo listening. What dimensions should it have in order to get optimum results with minimal problems to be solved using multi-subs and equalization? Such that I will not need to spend usd 10K to simulate my room later in order to optimize it?

My suggestion: build it 2 feet larger in each dimension (inside dimensions) than you want the final room to be, and then put in 8" spacers and 4" open-back sound absorbing panel everything. That spacing is very important, I do not mean "4" on the walls", very different effect. Ditto ceiling.

 

[j_j]

. But jj quoted Ethan referring to 6 dB peaks and 30 dB nulls, which in the given context will occur at different frequencies:

Or positions. It makes no sense to talk about different peaks and nulls at different frequencies and compare them in the context of handling a given room mode, after all, doesn't it?

If you have a 30dB notch somewhere, you have something close to a 6dB peak somewhere else AT THE SAME FREQUENCY. Goodness.

 

[KlausR.]

Originally Posted by KlausR.
But jj quoted Ethan referring to 6 dB peaks and 30 dB nulls, which in the given context will occur at different frequencies.

Or positions. It makes no sense to talk about different peaks and nulls at different frequencies and compare them in the context of handling a given room mode, after all, doesn't it?

I agree.

If you have a 30dB notch somewhere, you have something close to a 6dB peak somewhere else AT THE SAME FREQUENCY.

I agree.

BUT, when you listen to music, you usually do not walk around in the room. Maybe some do, but I for one don't. So I don't care about what happens elsewhere in the room, I care about what happens in the listening position and there I have nulls and peaks at different frequencies.

For home cinema seating areas obviously it's a different story.

Excuse me? You misquoted me and insisted I said something I didn't, and that makes me "disrespectful"? Really?

I don’t blame you for what you’ve said but how you've said it, no less, no more.

Klaus

 

[j_j]

I
BUT, when you listen to music, you usually do not walk around in the room. Maybe some do, but I for one don't. So I don't care about what happens elsewhere in the room,

Well, if you don't walk around a room, you're not actually understanding what the room is doing, well, unless you make a full soundfield measurement at your listening position, which will show a great deal of energy present even though the pressure is very small.

The point I am still trying to make is that a pressure zero at one point is not in any way an indication of less energy in a room. It is only less energy in a room if the speakers are not radiating energy at that frequency, which is not the usual problem, perhaps barring out of phase speakers, which I am hoping is not the issue. Even then, there isn't necessarily zero energy in the room.

 

[KlausR.]

Originally Posted by KlausR.

BUT, when you listen to music, you usually do not walk around in the room. Maybe some do, but I for one don't. So I don't care about what happens elsewhere in the room,

Well, if you don't walk around a room, you're not actually understanding what the room is doing, well, unless you make a full soundfield measurement at your listening position, which will show a great deal of energy present even though the pressure is very small.

I walk around the room while playing sine tones and hear what’s happening. I sit on my listening sofa while playing sine tones and hear what’s happening. I do the same while playing music and find that on the sofa only the 2nd order width mode is disturbingly excited on a few occasions. There’s a lot that I don’t understand fully, and I don’t have the necessary background in physics to figure it out myself.

As frequencies get higher, HRTF's convert part of the velocity field into pressure in the ear canal, and of course in a directionally sensitive fashion. How much is literally what determines your HRTF's.

Do you have a source that explains that in a manner comprehensive to the layman?

Klaus

 

[j_j]

Do you have a source that explains that in a manner comprehensive to the layman?

Klaus

**whew**

Um, wow, good question. Beranek is hardly a lay book. Let me ask around.

 

[KlausR.]

Um, wow, good question. Beranek is hardly a lay book. Let me ask around.

The 1986 edition is available in a nearby library, or would I need a more recent one? That library also has Howard/Angus 2009, Pressnitzer 2005, Fastl/Zwicker 1999. Fastl/Zwicker is on the shelf where I work (don't know which edition), what about Blauert?

Klaus

 

[j_j]

The 1986 edition is available in a nearby library, or would I need a more recent one? That library also has Howard/Angus 2009, Pressnitzer 2005, Fastl/Zwicker 1999. Fastl/Zwicker is on the shelf where I work (don't know which edition), what about Blauert?

Klaus

Jens has a good book on hearing, it's not exactly lay level, though. Sorry, that would be Blauert. The 1986 edition of Beranek might be useful, I'm not sure, it depends on which book (he has written several, each of which has several editions) it is. Fastl/Zwicker I would consider secondary for a variety of reasons.

 

[NMMark1962]

What a fascinating thread....though much of is is over my head...

Now, considering I will be using some large to very large speakers I will want plenty of space....

So what happens if I take one of the Golden Rules (1 x 1.6 x 2.33) and use a ceiling of of 16', I get a room that is 16' x 25.6' x 37.28' (imperial, not metric).....what happens if I were to keep the height at 16' but went to 27.6 x 38.9?? Would it all go to hell and my poor acoustical guy would, in a state of utter despair, run screaming off of a cliff??? Or, by slightly stretching the width and length a wee bit could the additions be easily corrected by traps/diffusers/absorption?? I am not saying I would do this (and my future audio guy says that these rules are not set in concrete and such numbers will be massaged when I select my brand of speakers) so I was just wondering if there is ever any flexibility in all of this.....

Cheers,

 

[amirm]

No matter what dimensions you pick, you will have low frequency room modes. By placing your speakers properly and adjusting your seating position, you can minimize them. BTW, your speakers cannot counter these effects. In low frequencies, the room is in control almost independently than speakers. Look at this measurement:

The graphs show what happens when you move the measurement mic from location to location. The speakers are obviously the same but we see wide variations in response due to the effects of the room below the transition frequencies of a few hundred hertz.

The ideal solution takes low frequencies away from main speakers so that they can be freely moved around to optimize bass response.

 

[j_j]

No matter what dimensions you pick, you will have low frequency room modes. By placing your speakers properly and adjusting your seating position, you can minimize them. BTW, your speakers cannot counter these effects. In low frequencies, the room is in control almost independently than speakers. Look at this measurement:

The graphs show what happens when you move the measurement mic from location to location. The speakers are obviously the same but we see wide variations in response due to the effects of the room below the transition frequencies of a few hundred hertz.

The ideal solution takes low frequencies away from main speakers so that they can be freely moved around to optimize bass response.

Amir, it would be interesting if you could swap mikes on that system, and make a measurement with the pressure (Omni) mike, and then make 3 more measurements in the same place, with a standard cardioid, pointed up, left, and straight ahead.

The cardioid would not be calibrated, but I think you'd find the results from the 3 directions interesting.

 

[KlausR.]

So what happens if I take one of the Golden Rules (1 x 1.6 x 2.33) and use a ceiling of 16', I get a room that is 16' x 25.6' x 37.28' (imperial, not metric).....what happens if I were to keep the height at 16' but went to 27.6 x 38.9??

What happens is that the distribution of mode frequencies is different, i.e. the frequencies are different as is the spacing between them. However, when you play music, how many modes are excited simultaneously, and to a degree that they disturb? One, two, more? Certainly not all of them as the mode calculator suggests, so the question arises: where’s the optimal place on the frequency scale for a single mode? So what’s the use of distributing 50 mode frequencies in an optimized manner when only one mode is driven at any one time?

Further, the mode calculator does not tell you anything about the level at which the mode is excited, and nothing about the level at which you perceive that mode. It does not tell you anything about happens in the time domain, how long it takes for that mode to decay.

So no, it won’t go all to hell, since you play music, not test tones, your room is built from real world stuff, not from infinitely rigid ones as the mode calculator assumes.

I’ve probably mentioned this in the thread: when building my room I determined the dimensions using an optimization criterion, the room was built slightly wider and it didn’t go all to hell.

…and my future audio guy says that these rules are not set in concrete and such numbers will be massaged when I select my brand of speakers;;;

I wonder why any well-educated mode would care about what brand of speaker is driving it.


@jj,

Checked Fastl/Zwicker and Blauert, and found nothing about this issue of “As frequencies get higher, HRTF's convert part of the velocity field into pressure in the ear canal…”


Klaus

 

[Ethan Winer]

what’s the use of distributing 50 mode frequencies in an optimized manner when only one mode is driven at any one time?

When two or more modes are close together, they combine to be even stronger than a single mode. This is the reason modes should be evenly distributed, and has nothing to do with how many different bass notes might be played in a given piece of music.

--Ethan

 

[j_j]

Checked Fastl/Zwicker and Blauert, and found nothing about this issue of “As frequencies get higher, HRTF's convert part of the velocity field into pressure in the ear canal…”


Klaus

Hm, it's bog-basic acoustics, they might not address HRTF's in specific, but look for diffusion and diffraction, then. Look for interference patterns generated by spheres, ellipses, etc.