Header image  
by Ralph Glasgal
line decor
Home Tutorials Tech
Kudos and
Demos Bio Free Ambio
Glossary The Home
Concert Hall
Rec Engineers
FAQ/Forum Links Contact us
line decor

Audio Engineering Society

Presented at the 99th Convention
October 6-9, 1995 New York, NY



The Synthesis of Concert-Hall Sound Fields in the Home

by Ralph Glasgal


We have seen that the ear is extraordinarily sensitive to horizontal angular position. We have also shown that the presence of the acoustical interaural crosstalk which is present in any stereophonic listening triangle sets up a pattern of erroneous peaks and nulls, which both defines and distorts the imaging quality of any particular installation. Even at frequencies below 500 Hz the ear can hear postion changes based on a 40 µsec time shift between the ears. This makes high-quality stereo systems susceptible to any small tweaks that differentially or absolutely modify peaks and nulls. The comparison of amplffiers, loudspeakers, cables, and phono cartridges based on image characteristics such as definition, stage width or air, is not a very productive undertaking. In the presence of stereo crosstalk the always imperfect image is very susceptible to very small changes in head position or differential delay, phase, or intensity; listeners will always find it difficult to decide whether or not such a change is for the better, since, even after the change, the image is still imperfect but in a different way.

One way around this is to do comparative listening tests of such equipment, using just one channel, and if you can bear it, just one ear. The other possibility is to compare or audition equipment with the crosstalk eliminator in place in a well-damped room. Evaluating equipment based on apparent stage width, image palpability, etc., is not technically supportable. A reviewer may be able to achieve a superb sound-stage image by tweaking amplifier, cable, speaker, and speaker position but another listener with different-shaped pinnae and different head size will have a different interaural crosstalk comb-filter pattern, and thus will not sense the same sound image. Finally, this same system if set up again with different speaker spacing or angle to a listener and perhaps a two-inch difference in path length to one ear, is likely to project an image quite different from the reviewer's original setup.


The presence of the barrier makes it difficult to add video to the system so as to enjoy opera or concerts on laser disk. Howwer, I have recently had much success using virtual-reality goggles from I-Glasses that do not interfere with the ears any more than reading glasses do. Another approach is to hang a small LCD panel monitor, like a picture, on the edge of the barrier. The closeness of the picture in this case may cause eyestrain, or require reading glasses.

The new AC-3 discrete five-channel surround-sound system could be used to provide three ambience signals in addition to the main left and right signals. Note, however, that the crosstalk barrier is still required, and that at least six independent ambience signals are desirable. In general a home synthesizer will always be cheaper, more flexible, and closer to audiophile standards than any new Ambiophonics recording standard. It is hoped that as the Ambiophonics method becomes part of the audiophile lexicon, easier-to-use and lower-cost ambience synthesizers will make their appearance and that recording engineers will make recordings that take into account Ambiophonics characteristics and thus move the dummy head further back from the stage, minimize the amount of rear hall reverberation picked up, and if the direct sound stage angle is wide, use a dummy-head with pinnae so as to compensate at least a little for the central position of the main loudspeakers in the Ambiophonics listening room. Finally, the industry should begin to think about a standard preamble for digital recordings or broadcasts that would automatically set the ambience synthesizer parameters to values that the recording engineer favors. The listener could still tweak to taste but the recording engineer knows precisely what the recording site was like, and is in an excellent position to make good choices.


The Ambiophonics configuration has been auditioned by hundreds of listeners many of them experienced audiophiles, who have agreed that the Ambiophonics paradigm for reproducing music from ordinary two-channel stereophonic (or matrixed surround-sound recordings) is capable of consistently creating a rock-solid sound image that extends far beyond the right and left positions of the front loudspeakers, usually extending to 120° or more. It has been established that the use of any one of several commercially available absorption barriers, if used to prevent interaural crosstalk from the front loudspeakers at the listening position permits the loudspeakers to be moved to a position almost directly in front of the home listener not only without compromising the perceived width of the stage image, but actually increasing it to more than 60° beyond the speakers. This positioning increases the binaural nature of the reproduction by eliminating the pinna-response errors caused by widely spaced loudspeakers and allows the speakers to mimic the usual microphone aspect. Ambiophonics also demonstrates that two-channel recordings, particularly those made with dummy heads or head-spaced microphones, provide sufficient data to the ear-brain system to sense very precise stage positions with depth and, as in other situations, move one's head and focus on a particular stage sound source.

Finally, it has been demonstrated that synthetic ambience and reverberation signals, if propagated in a properly treated room if launched from the proper direction so as to stimulate the correct pinna response, if reasonably matched to the recording, and if delivered with as low an interaural correlation value as possible, then almost all listeners will sense that they are in the same space as the performers. Furthermore, it is a basic tenet of Ambiophonics that the generation of early reflections and reverberant fields will always be better done by personal computers or digital signal processors than by multichannel recordings. The advantages of the home electronic generation of the ambient field are that no new multichannel recording standard need be established, that the size and character of the home concert hall can be changed at will to suit one's tastes, that non-multitrack or even monophonic recordings will sound just as good or even better than any prospective multichannel system, and that a one-time investment in ambience-generating software or hardware will be less expensive than paying a premium for every multichannel record purchased.


Y. Ando, "Concert Hall Acoustics," Springer Verlag, 1984

Timothy M. Bock & D. B. Keele Jr., "The Effects of Interaural Crosstalk on Stereo Reproduction"
presented at the 8lst AES Convention, Nov., 1986 [Preprint #2420-A & 2420-B]

Ralph Glasgal & Keith Yates, "Ambiophonics-Beyond Surround Sound to Virtual Sonic Reality," Ambiophonics Institute, Northvale, N. J., June, 1995

Ralph Glasgal, "The Domestic Concert Hall," Stereophile, July, 1988.

Ralph Glasgal "The Domestic Concert Hall, Reviewed" The Audiophile Voice, vol.2, Nov., 1994

Møller, Sørensen, Hammershøi, Jansen "Head Related Transfer Functions of Human Subjects," Journal of the Audio Engineering Society, Vol. 43, No.5, May, 1995

Snow, Blumlein, Moir, Venneulen, Clegg, et al., "Stereophonic Technique," AES, 1986