Psychoacoustics of Room Reflections

amirm · Oct 12, 2012

Phelonious Ponk said:
Amir, this is one area where I would join the subjectivists and say "trust your ears." Anechoic rooms sound bad, and they start sounding bad before they get to absolutely anechoic. Of course most of us don't get much of a chance to hear them, but there are a couple of reasonable substitutes -- have some stand mount speakers you are very familiar with? Take them outside and play them into an open space. Dry as unbuttered toast. Some in-ear monitors, with very flat FR just don't sound right either. Which is why, if you look at FR charts on headroom, for example, you'll see substantial amounts of "room gain" built into the very best headphones on the market. I love the confirmation all your good work provides, but I'm not surprised by the results. And it's good news for people who don't want to live in domestic spaces that look like studio control rooms.

Tim

There are surprising results though. A lot of which I have not covered in that article. The bit that I did with Clark's test is the start of that. Note that a reflector held vertically produced almost no effect whereas rotating it 90 degrees, it produced a negative result. This shows that floor/table/console reflections are bad indeed. Further research into that shows that it changes timbre but only if the reflection frequency is 500 Hz and up. This is very fortuitous since it means we can get rid of the floor reflections with pretty thin material such as thick carpet and padding. Imagine if we had to put 4 inches of fiberglass there!!!

Here is a useful graph from Clark paper:

The top three shows the Impulse Response of three of the scenarios. As we see, the response is nearly identical. There is the initial impulse followed by another strong "reflection." The two graphs on the bottom right show what each ear is hearing. We notice that the frequency is different since the distance to the reflection changes depending on which ear is closer to it (the graph to the bottom left shows them overlaid with one of them boosted artificially). This difference causes the brain to tune out a lot of the notches in the comb filter.

rbbert · Oct 12, 2012

OK, I have a further question to which I don't find a clear answer: what about diffusion at the first reflection points? I know there was a review in Stereophile recently by Eric Lichte about the benefits of this, and Art Noxon of ASC has been a proponent of this approach for awhile as well.

microstrip · Oct 13, 2012

rbbert said:
OK, I have a further question to which I don't find a clear answer: what about diffusion at the first reflection points? I know there was a review in Stereophile recently by Eric Lichte about the benefits of this, and Art Noxon of ASC has been a proponent of this approach for awhile as well.

As far as I know the big issue with diffusors is that most of them are not true diffusors.

as they have a very limited band of operation and some also behave poorly at low incidence angles. In order to operate down to 300 Hz a diffusor needs to be very deep. But if you have the space and good quality diffusors it seems they can be an interesting possibility.

amirm · Oct 13, 2012

rbbert said:
OK, I have a further question to which I don't find a clear answer: what about diffusion at the first reflection points? I know there was a review in Stereophile recently by Eric Lichte about the benefits of this, and Art Noxon of ASC has been a proponent of this approach for awhile as well.

A diffuser is another approach to the same theme of absorbing the reflection with the difference being that the energy is kept in the room. In that sense it still goes against the research of leaving the side walls fully reflective. If you are going to deploy them, then you need to do as Microstrip says which is to make sure they are broadband. This can be a depth of 8 to 12 inches.

microstrip · Oct 13, 2012

I first met diffusors in the classical article of Keith Yates about diffusion in Stereophile some decades ago:

http://www.keithyates.com/matterofdiffusion.htm

Unhappily at that time there were no free plans of QRD diffusors in the net, and my experiences with poorly designed home made devices was not very successful. Only almost fifteen later I was able to buy a lot of RPG diffractals from a closing shop. I still use them in the front wall of my room. Their depth is 9-1/8'' .

Soundminded · Oct 13, 2012

amirm said:
OK, here at long last, is my article in the last issue of Widescreen Review, summarizing research and listening tests that says our intuition and "common wisdom" about how the room impacts the sound of our speaker is wrong. The focus of this article is on frequencies above the transition (200 to 400 Hz) and relates to the so called "early" or "strong" reflections. It does NOT deal with mid to late reflections. That will be another article . Unlike my previous article on low frequency optimizations, this article is not an end to end tutorial but rather, sets the stage for the science that governs that. Unless one accepts and comes to term to the view of acoustics here, the other bit won't make sense.

So as always, please let me know what you think across the board.

Perceptual Effects of Room Reflections, AKA, "It is not Simple!": http://www.madronadigital.com/Library/RoomReflections.html

I've been experimenting with room reflections since just about as long as I can remember. Look at the size and geometry of a concert hall and compare it to your listening room. There is no comparison. Anyone who thinks they can get a pair of speakers playing a 2 channel recording using microphones placed close to the musicians to sound like a symphony orchestra in a room hundreds of times the size of his own is just kidding himself. If that's to be done, it will have to happen another way. So what can be done? At best you can reproduce the sounds of musical instruments as they would be heard in playing in your room. That limits the value of conventional stereo systems to soloists or small groups at best But not so fast. To do that, the sounds have to arrive at your ears the same way they would from those instruments. That means from many directions in rapid succession and with nearly the same FR. That won't happen for any commercial speaker I've ever seen. Vectors are all wrong. The speakers for the most part are not only point sources, they beam most of their sound directly at you, in fact almost always all of the energy in the highest octave is beamed at you. That means the reflected sound has no high frequency energy. And it's way out of proportion because over much of tis range, speaker systems are fairly directional. What about the one effort that tried to vector sound more naturally at least in the horizontal axis, the infamous B9? Well it produced almost no energy in the top octave and what little it did, at least forward was even more directional owing to its large midwoofers. And it did not compensate for spectral reflection distortions. Could it be re-engineered? Yes, it took two tries, the second spanning four years even knowing what the model required. And it still has to be equalized for each recording individually to sound accurate. To these critical ears, that's a long tedious process that requires many repeated efforts to get it right. When it is right, it works very well...in this limited context.

Wayne A. Pflughaupt · Oct 20, 2012

Kudos on an excellent article, Amir.

Regards,
Wayne A. Pflughaupt

LenWhite · Oct 20, 2012

amirm said:
OK, here at long last, is my article in the last issue of Widescreen Review, summarizing research and listening tests that says our intuition and "common wisdom" about how the room impacts the sound of our speaker is wrong. The focus of this article is on frequencies above the transition (200 to 400 Hz) and relates to the so called "early" or "strong" reflections. It does NOT deal with mid to late reflections. That will be another article . Unlike my previous article on low frequency optimizations, this article is not an end to end tutorial but rather, sets the stage for the science that governs that. Unless one accepts and comes to term to the view of acoustics here, the other bit won't make sense.

Thanks for a very interesting article Amir.

Your article states absorbing early reflections is actually counter productive because in "normal" listening rooms early reflections actually tend widen the soundstage. A room sans acoustic treatment tends to sound "edgy" to my ears, and I can definitely hear "echos" in untreated rooms especially when I clap my hands together. I've always thought those types of "echos" smeared sound reproduction. But it has been a very long time since I've listened to my audio system without absorbers on the sidewalls. I need to try that I guess

I'm very interested in your opinion regarding the other walls in a stereo oriented listening room. I've also always thought home theater and stereo listening rooms had very different acoustic objectives. I've tried to optimize my room for stereo listening based on my understanding it should be a more lively room than home theater, but also believing acoustic treatment is still a key component in great sound reproduction.

Your article also indicates many (most) people are measuring frequency and time incorrectly in listening rooms. I'm using difussors on the front wall and ceilings, and absorbers on the side and rear walls, except for the area 5' directly behind the listening position. And I believe my listening room and audio system setup provides great dynamics, articulation, timbre, and imaging. But I've never had my listening room "professionally" measured, perhaps I should

I appreciate your expertise.

rbbert · Oct 21, 2012

I'm a little confused by all this; I don't have a lot of treatment at sidewall first reflection points, but the image seems both wider and deeper when I have a little absorption there compared to bare wall (I just spent some time putting the absorption in and out)?

amirm · Oct 21, 2012

LenWhite said:
Thanks for a very interesting article Amir.

You are welcome. Thanks for reading and commenting.

I'm very interested in your opinion regarding the other walls in a stereo oriented listening room. I've also always thought home theater and stereo listening rooms had very different acoustic objectives. I've tried to optimize my room for stereo listening based on my understanding it should be a more lively room than home theater, but also believing acoustic treatment is still a key component in great sound reproduction.

The role of other surfaces is a bit complicated in that they can contribute to other things. The article here dealt with early/strong reflections. These occur on the side walls, floor and ceiling. The article covers how the side walls can be beneficial in widening the point source of the speaker. Turns out this is more useful in stereo than multichannel as the latter has so many direct clues coming from other speakers that this aspect is not as necessary.

The floor reflections are not a good thing to have. So they should be absorbed. Fortunately, what they do is change timbre and their effect is dominated above 500 Hz. This means that a reasonable thick carpet with good padding does the job. Ceiling is not as critical but general advice is to get rid of direct reflections. The common solution is a diffuser there. Back wall also falls in the area of non-beneficial reflections. So if you need additional absorption of mid to late reflections, that is a good place to put them. This is the bit I mentioned above. Mid to late reflections should be controlled as too much of them make the room too live. Too little is not good either. If you can measure it, you want to aim between 0.3 and 0.4 in the 500 Hz region. If it is higher, then go ahead and put more absorption on the back wall. Front wall can likewise be treated. Put thick ones there if you can (4 inches or better) so that they are broadband. If using diffusers, they need to be at least 8 inches.

Your article also indicates many (most) people are measuring frequency and time incorrectly in listening rooms. I'm using difussors on the front wall and ceilings, and absorbers on the side and rear walls, except for the area 5' directly behind the listening position. And I believe my listening room and audio system setup provides great dynamics, articulation, timbre, and imaging. But I've never had my listening room "professionally" measured, perhaps I should

It is very easy to get confused. As three acoustics people and they give you three different advice

. What I am mentioning is based on decades of actual research and listening tests rather than what seems right. Your front and ceiling are fine if they are deep diffusers. If the side wall absorbers are on a different spot than first reflection, that is probably fine too as long as they are thick.

I appreciate your expertise.

I hope I can live up to the expectation!

Longer term I will write an article on the points above. The much longer version of the same is in Dr. Toole's book of acoustics and speakers if you can't wait that long (http://www.amazon.com/Sound-Reprodu...F8&qid=1350863040&sr=8-1&keywords=toole+floyd).

amirm · Oct 21, 2012

rbbert said:
I'm a little confused by all this; I don't have a lot of treatment at sidewall first reflection points, but the image seems both wider and deeper when I have a little absorption there compared to bare wall (I just spent some time putting the absorption in and out)?

My hypothesis is that they are helping reduce mid to late reflections (a different issue than first/strong reflections covered in my article). To the extent you have a bare room and just a chair and carpet, you need additional absorption. If you know how to measure RT60, you can determine if this is needed and how much. Only a fully furnished room has enough of absorption. Dedicated rooms need fair bit of them. So you adding those absorbers may have accomplished this goal.

amirm · Oct 21, 2012

Wayne A. Pflughaupt said:
Kudos on an excellent article, Amir.

Regards,
Wayne A. Pflughaupt

Thanks partner

.

microstrip · Oct 21, 2012

Amir,

Do you have an article about the effect of ceiling/floor reflections and their treatment?

amirm · Oct 21, 2012

microstrip said:
Amir,

Do you have an article about the effect of ceiling/floor reflections and their treatment?

I have not written one. I did post in another forum the relevant research. Here it is again for ceiling:

From psychoacoustics point of view, ceiling reflections are a bit of an enigma. One would normally expect the ear/brain to no be hardly sensitive to them as the same sound is reached both ears unlike side reflections (i.e. what happens due to front wall reflections). Listening tests by both Harman and Rakerd, et al. however show that the threshold of detection is actually the same as side reflections. Why the brain is more sensitive when the vertical angle is increased is not very clear. This is the hypothesis by Rakerd paper: (HP means horizontal plane/side reflections and MSP is median sagittal plane/Ceiling reflections):

"The present psychophysical findings of equivalence in echo suppression for the HP and the MSP can be interpreted two ways. It is possible that the suppression occurs at a neural processing site that is indifferent to source location. Alternatively, the findings may be seen as consistent with recent neurophysiological evidence ~Litovsky and Yin, 1994; Yin and Litovsky, 1994; Litovsky et al., 1997! for an echo suppression mechanism mediated by higher auditory centers where binaural and spectral cues to location are combined."

While echo suppression is the same as side reflections, the audible effect is limited to timbre change only. I.e. no increase occurs as far as speciousness and such (i.e. it is not a beneficial reflection). Dr. Toole's optional recommendation is to use a diffuser to push the reflections to the side walls. Keith Yates is also a fan of diffusion in the ceiling. Tony Gremoni suggest an absorber although he leaves the option for diffuser.

amirm · Oct 21, 2012

And for floor reflections from Dr. Toole:

"Looking at the absorption coefficients used in modeling the floor reflection (Bech, 1996, Table II) reveals that the simulated floor was significantly more reflective than would be the case if it had been covered by a conventional clipped pile carpet on a felt underlay. Further investigations revealed that the detection was based mainly on sounds in the 500 Hz–2 kHz range, meaning that ordinary room furnishings are likely to be highly effective at reducing first reflections below threshold, even for the more demanding signal: broadband pink noise (see Section 21.3).

In terms of spatial aspects, Bech (1998) concluded that those sounds above ?2 kHz contributed to audibility and that “only the first-order floor reflection will contribute to the spatial aspects.” The effect was not large, and, as before, speech was less revealing than broadband noise. Again, this is a case where a good carpet and underlay would appear to be sufficient to eliminate any audible effect. See Figure 21.3 for data on the acoustical performance of floor coverings.”

Here is figure 21.3:

We see that with just a half inch of padding, we get to have the type of absorption we need a 500 Hz and above.

c1ferrari · Oct 21, 2012

Acoustic Parameters to Be Considered...

Hello Amir,

Would you have any comment on acoustic parameters to be considered for an optimal tracking room, optimal mixing room, and optimal mastering room, i.e., three separate rooms/spaces with three different functions

Thanks

JackD201 · Oct 22, 2012

microstrip said:
As far as I know the big issue with diffusors is that most of them are not true diffusors. as they have a very limited band of operation and some also behave poorly at low incidence angles. In order to operate down to 300 Hz a diffusor needs to be very deep. But if you have the space and good quality diffusors it seems they can be an interesting possibility.

It's an issue if one tries to leverage the absorptive qualities of a diffusor's design. When chosen and placed properly their prime (no pun intended

) purpose of even coverage is the greater benefit.

rbbert · Oct 22, 2012

amirm said:
My hypothesis is that they are helping reduce mid to late reflections (a different issue than first/strong reflections covered in my article). To the extent you have a bare room and just a chair and carpet, you need additional absorption. If you know how to measure RT60, you can determine if this is needed and how much. Only a fully furnished room has enough of absorption. Dedicated rooms need fair bit of them. So you adding those absorbers may have accomplished this goal.

This almost sounds like saying a typically furnished (i.e., non-dedicated listening) room needs no treatment at all (except perhaps bass trapping?)

microstrip · Oct 22, 2012

amirm said:
I have not written one. I did post in another forum the relevant research. Here it is again for ceiling:

Great. I see now that I should have enlarged my search horizon to the avsforum.

Very thoughtful debates!

theguesswho · Oct 22, 2012

"This almost sounds like saying a typically furnished (i.e., non-dedicated listening) room needs no treatment at all (except perhaps bass trapping?) " said rrbert.

Yes,dont you think so? I think that is what Amir is saying. Think about it, we have been listening to music in rooms for hundreds if not thousands of years, did we need acoustic treatment? No, this is a recent phenomena that is bolsterd by MANUFACTURERS of "room treatment/acoustics! We dont need it in a typical room see this

http://www.linkwitzlab.com/frontiers.htm#A

Where Siegfried Linkwitz said "Most of the research into audibility of reflections has been performed under anechoic conditions. Detection thresholds and phantom image shifts for a single reflection versus delay time and amplitude have been studied extensively. But, the results cannot necessarily be translated to binaural listening in reverberant spaces. Nevertheless, the prevailing school of thought amongst audiophiles holds all reflections to be detrimental. The prescription is to either absorb or diffuse reflections as much as possible. A room with only hard reflective surfaces and with no furnishings, having echoes and long reverberation time, is clearly not an ideal playback environment, but nor is an overstuffed room, full of absorbing materials and resembling an anechoic chamber."

Wendell

Psychoacoustics of Room Reflections

Banned

Well-Known Member

VIP/Donor

Banned

VIP/Donor

New Member

Well-Known Member

Well-Known Member

Well-Known Member

Banned

Banned

Banned

VIP/Donor

Banned

Banned

Member Sponsor & WBF Founding Member

WBF Founding Member

Well-Known Member

VIP/Donor

New Member

Similar threads