AES Journal Forum

Reflecting on Reflections

(Subscribe to this discussion)

Document Thumbnail

[Feature] Spaces for live performances are part of the art, enriching timbres, embellishing direction and space, and enveloping the audience. These spaces are all different from each other, but designed to meet certain acoustical criteria related to the kinds of performances for which they are intended: opera, orchestral classics of different periods, and so on. Creating a strongly reflective, diffuse, sound field that conveys the unamplified sounds from live performers to the farthest seats in concert halls without corrupting the temporal details in the performance is a fine balancing act. This is the domain of traditional acoustics. This is Sabine’s “place.” But things change.

JAES Volume 62 Issue 6 pp. 454-455; June 2014
Publication Date:

Click to purchase paper as a non-member or you can login as an AES member to see more options.

(Comment on this feature)

Comments on this feature

Siegfried Linkwitz
Siegfried Linkwitz

Comment posted July 6, 2014 @ 16:06:36 UTC (Comment permalink)

It is about time that the audio industry and academia address the loudspeaker and room compatibility issue for domestic size reverberant spaces and recognize it as just one more problem area in the transmission path from microphone membrane to eardrum. The brain takes over behind the ear drums, creating the perception of an acoustic event in some form of 3D space. The perceived scene is spatially distorted due to the directional characteristics of microphones and their distances to sound sources, and due to the directional characteristics of loudspeakers and their distance and azimuth to the listener. The microphone output signals contain a limited and specific view of the original acoustic scene. On playback the loudspeakers illuminate the listening room. The resulting eardrum signals consist of superimposed streams of air pressure variations arriving from different directions.  With two ears, a movable head, a brain and a mind, evolution has formed a superbly capable perceptual apparatus, which automatically sorts a plurality of sound streams for direction, distance, meaning and for focusing attention.  Loudspeakers and room (also the room behind the speaker membrane) must not provide misleading cues to a listener's brain in order to fully hear the information contained in the microphone signal streams and to instinctively withdraw attention from loudspeakers and listening room. The result is a convincing auditory illusion.

The job can be done. It has been done. A simple stereo system can render in an ordinary room a more convincing 3D illusion than conventional multi-channel surround setups, because it can put perception at ease.
Why has the audio industry not caught on other than trying to put patches on the problem for decades?
Siegfried Linkwitz -

Default Avatar
Bob Olhsson

Comment posted February 23, 2015 @ 15:19:25 UTC (Comment permalink)

I'm pleased to see Dr. Toole talking about what the folks at RCA learned about the importance of early reflection response during the '40s! Combine this with Richard Heyser's work and we emerge from 40 years in the dark ages of listening room design!

Default Avatar
Author Response
Floyd Toole

Comment posted February 25, 2015 @ 16:58:58 UTC (Comment permalink)

Bob.  RCA in the '40s.  I can only imagine you are talking about the legendary Harry Olson - any relation, the names are remarkably similar?  Yes, Harry contributed a lot of solid engineering, good sense and logical thought to the audio industry, much of it ignored or  forgotten it seems.  I am frequently humbled by the insights of our predecessors, who lacked the high technology measurements and analytical tools we have.  There is a lot of old fashioned folklore that needs to be purged from our industry.  Science is hard work, opinions come easily.  Thanks for your comment.

Default Avatar
Bob Olhsson

Comment posted February 27, 2015 @ 16:45:17 UTC (Comment permalink)

Prior to the '50s Bell Labs and RCA paid some of the highest engineering salaries in the world and it showed in what they accomplished. It's too bad that level of investment is gone. Instead of science we now have mostly multiple myths being argued.

The acoustical treatment for our little studio at Motown was designed by RCA. It was designed to have flat, uncolored early reflections. I found that adhering to this principle has always achieved the best results in both recording studios and listening rooms. It's even logical that early reflections will always color perception and there is no way to eliminate them short of an anechoic chamber.

Default Avatar
Gary Eickmeier

Comment posted January 6, 2017 @ 19:03:17 UTC (Comment permalink)

Mr. Olhsson's comments are refreshingly correct. Dr. Toole's final 3 paragraphs are also correct about the 3 choices on what to do with off axis sound. The final choice, reflecting the sound from hard flat surfaces is the correct one, and agrees with Olhsson. Let me explain.

For the entire stereo era and especially with advice given to audiophiles and engineers today, stereo theory has been shamefully incorrect, leading to bad practices in speaker manufacture and advice on their use. The graphics at the bottom of Dr. Toole's article illustrate the problem.
The one on the left shows the live sound situation and the one on the right the reproduction. I agree that the multi channel illustration would be better than just two channel stereo, but the basic two channel part is where I am going. It seems that we have been fed a lot of information that the direct sound from the two speakers is primary and all important. We have been told to dampen or eliminate the reflected sound output from all speakers. This is based on an incorrect stereo theory that the idea is to direct the two recorded channels to our two ears, and then we will "hear" what is contained in the recording in all its glory.
What is forgotten in this simplistic theory is that the all important spatial nature of the live sound will then be forced to arrive at the hapless listener's ears from only those two points in space, and not the multiplicity of incident angles that were recorded. The spatial nature of the live sound has been changed to a high direct field from two points only.
My claim is that this "head related" theory is fundamentally incorrect. What we are doing in stereo is not this "two speakers, two ears" head related theory but rather a field-type system in which the object is to reproduce sound fields in rooms, not signals for the ears. You do this by bringing the speakers out from the reflecting surfaces and directing the majority of their output in the reflecting directions and just a small amount in the direct field from the speakers themselves. By using specular reflectivity from the walls we retain the same frequency response as the actual speakers and incur an appropriate delay of the reflected sound to remain within the fusion time, and reproduce the spatial nature of the original. 
The paradigm now becomes the image model of the live vs the reproduction sound fields. The focus is now how closely these two fields match. I call it an image model theory for stereophonic sound, which I would hope will replace the head related direct sound so-called "accuracy" theory of the past. I am doing it in my home and it gives a 3-dimensional model of what is contained in the recordings with great depth, spaciousness, and realism of imaging that is much superior to any direct sound system I have ever heard.

Default Avatar
Bob Olhsson

Comment posted January 9, 2017 @ 14:12:56 UTC (Comment permalink)

Flat walls don’t necessarily reflect a flat response and can cause flutter-echo which is a distraction. Acoustical treatment and diffusion can be used to flatten the reflection response.

Default Avatar
Gary Eickmeier

Comment posted January 10, 2017 @ 18:57:33 UTC (Comment permalink)

There might be flutter echo if you were doing it in a bare room, but not in a room designed for audio - either production or reproduction. For example, my room is 20 ft wide by 30 ft long and has some diffusers in the back half on the side walls, plus absorbent furniture at the listening position and plush throw rugs on all floors, and a cathedral ceiling. The RT60 is about 500 ms, about right for this size room. The walls near the speakers at the front of the room are flat wallboard done purposely to create a huge lattice of direct and reflected image sources. What all this means is that the first reflections are specular and occur only once, then are absorbed normally, the same as any other system. All that changes is the spatial characteristic, leaving the temporal unaffected except for what is contained in the recording. Doing the spatial by reflection like this is the only way to be able to adjust the direct to reflected ratio and make it more like the live model. It's a whole different universe with different rules. Paper to follow - hopefully!

Subscribe to this discussion

RSS Feed To be notified of new comments on this feature you can subscribe to this RSS feed. Forum users should login to see additional options.

Join this discussion!

If you would like to contribute to the discussion about this feature and are an AES member then you can login here:

If you are not yet an AES member and have something important to say about this feature then we urge you to join the AES today and make your voice heard. You can join online today by clicking here.

AES - Audio Engineering Society