[Continued from Part One. . . .]
There are two other stereo miking techniques I've heard which, to my ears, result in a fairly accurate spatial perspective, one where playback yields a fairly convincing replication of the perspective the listener would have heard from the recording mike position in the recording hall. One is the binaural method of using tiny condenser mikes inside a dummy head/shoulders/pinna baffle, with listening done through headphones. Most listeners agree that remarkably convincing spatial perspectives can be achieved with this technique when playback is through headphones. I know that others disagree, but I also find such recordings fairly convincing from a spatial perspective even when played back through loudspeakers if the speakers are placed so that the subtended angle is less than 60 degrees, such as is usually the case when the speakers are set up by the rule of thirds pointing down the long dimension of a rectangular room. You be the judge; here is the site where you can order binaural recordings.
What is needed is a technique which most can agree offers binaural-quality soundfield with loudspeaker playback. I’d like to suggest that the Dimensional Stereo Microphone (DSM) technique being marketed by Sonic Studios, which is arguably a variation of the binaural technique, is just such a system. Truly remarkable results can be achieved when tiny listener-worn omnidirectional condenser mikes (usually mounted on eyeglasses) are played back either through headphones or loudspeakers. If the recording is intended for loudspeaker playback, the microphones are simply moved forward of the ears along the glasses earpieces toward the lenses. See, for example, http://www.sonicstudios.com/dsm.htm and http://www.sonicstudios.com/anarchy.htm. Such recordings can sound amazingly realistic even though the quality of the available mikes for this application is a bit questionable.
One obvious problem with this technique is that the listener’s own personal recordings while wearing the mike probably map better to that listener’s ears than to others' ears. Another set of problems are caused by the self-noise and perspective changes caused by even small movements of the listener/recordist while recording the performance, movements which cannot be compensated for upon playback, even by the same listener who recorded the event.
Great results with this technique can also apparently be accomplished using relatively inexpensive dummy heads as baffles. What you lose to the “averaging” involved with such dummy heads, you gain back in stability of image and repeatability of results. Sample recording files using this technique with a dummy head as a baffle are available at http://www.sonicstudios.com/mp3.htm. These files sound remarkably live even when streamed off the Internet through my cable modem connection.
There are, of course, very few commercial examples of such recordings, this technique having been developed primarily for making "bootleg" recordings of live events from the audience. Many of the products and accessories available from Sonic Studios are aimed at addressing problems created by the need for stealth and ultimate portability of the recording system. These problems would evaporate in the context of serious professional recording equipment. The DSM technique thus strikes me as a potentially profitable avenue for further research since such mike arrays seem to be the best ones available for mapping the recording perspective fairly well to the listening room perspective with two-channel stereo. I would imagine that the high-quality artificial head and shoulders mannequins used for the best binaural recordings would work well with the DSM technique if the mannequins are simply fitted with glasses or other means for mounting the mikes a bit forward of the dummy's ears. Combine this with extremely high quality miniature condenser mikes and we may really have "something to write home about," as they say.
More Channels Needed
But most theorists agree that for truly accurate three-dimensional spatial mapping of sonic event, more than two channels are necessary. The Sonic Studios folks have experimental surround-sound models available where the head is equipped with at least four mikes. See http://www.sonicstudios.com/index.htm and scroll down.
Alternatively, I suggest further research into creating microphone arrays which can move beyond the Blumlein technique's accuracy in capturing spatial perspective within the 90-degree included angle of the mikes. While Blumlein obviously captures sound from the sides, behind, and above, it does not map such sounds accurately onto the soundstage as reproduced by two stereo speakers. This may be the reason why most practitioners place even Blumlein mike arrays closer than the perspective goal would indicate: the room sound is captured from all around the mike array, but with two-channel stereo playback, this extra reverberation comes from the two stereo speakers up front along with the direct sound, adding an extra dollop of concert hall reverberation to the direct sound which tends to make the source sound further away because of the increased ratio of reverberant sound coming from the same general direction as the direct sound of the instruments.
Perhaps all that is needed is for the information captured by a single Blumlein pair to be extracted and distributed among four or more channels around the listening position. If real height information is desired, a second Blumlein pair aimed between floor and ceiling might capture this information. There should be some way of combining the outputs of one or more Blumlein stereo-pairs aimed in different directions to accurately map a full 360-degree soundfield, at least when played back through some reasonable number of channels. I will leave it to the mathematicians and other theorists to figure out how many channels are necessary and how the microphone pairs should be oriented and their outputs summed or subtracted to yield such a result. Maybe the solution is as "simple" as a resurrection and enhancement of the Ambisonics technique. Note that REG used some sort (he hasn't said exactly what he did) of Blumlein-related technique to produce very convincing rear-channel surround on the Water Lily Mahler 5th Symphony SACD recording he was involved with.
I do know that with DVD, SACD, and Blu-ray, we now have commercial media which can give us at least five channels to play with in an attempt to accurately recreate a soundfield. Whether five or seven is enough to unambiguously map the soundfield from recording hall to listening room I do not know, but we now have media which can at least get us closer to the absolute, if that if where we want to go.
I suspect, however, that the vast majority of commercial recordings will never go there. Perhaps I am overly pessimistic about such things, but it seems quite likely that commercial releases on surround-sound-capable media will move in the direction of an attempt to put the listener in the middle of the musicians, and that the limitation of the extra channels to ambient effects will be deemed too subtle and “wasteful” of all those extra channels, especially given the bleed-over in practice and expectations from the established home theater market, where such sonic shenanigans are already the norm.
“They Are Here” as a Parlor Trick
"They are here" is a lot easier goal to attain, but is really on the order of a parlor trick, especially for a single instrument. Record any instrument with good mikes in a single-point stereo array from quite close up and you will effectively ignore or erase the room sound of the recording venue. Play that recording back with flat speakers (well, maybe a bit of a HF rolloff to compensate for the very close mike positioning in the recording), with reasonable dispersion and low distortion, at the right volume level in a relatively live room and--Presto!--you have "they are here" in the listening room, since the listening room acoustics don't fight with recorded acoustics and thus the listening room's sound takes over almost completely.
Acoustic Research once used a twist on this technique to confuse listeners in a blind test in a concert hall as to whether the orchestra was playing or whether the sound was coming from a battery of their speakers set up on stage in front of the musicians. It worked because the recording of that orchestra was made outdoors or otherwise anechoically. When the recording was played back in the hall, the concert hall acoustics so swamped the direct sound from the speakers that listeners supposedly couldn't tell the difference.
As others have noted, "they are here" really doesn't make literal sense for large groups anyway since you could not physically get an orchestra into a listening room.
Listeners who prefer the "they are here" goal usually do so because, in this day of movies and home theater, many crave the overt aural stimulation that immediacy and high volume yield: a surreal or "better than real" sound that "sounds good" to them. I can't say I totally blame them; I like the added immediacy of close seats even for large orchestral groups and actively dislike listening to small acoustic instrumental groups from far back in a hall. There is something to be said for higher volume, more immediacy, and exaggerated staging and envelopment to make up for what you lose in not being able to actually see the musicians before you.
[Concluded in Part Three. . . .]
There are two other stereo miking techniques I've heard which, to my ears, result in a fairly accurate spatial perspective, one where playback yields a fairly convincing replication of the perspective the listener would have heard from the recording mike position in the recording hall. One is the binaural method of using tiny condenser mikes inside a dummy head/shoulders/pinna baffle, with listening done through headphones. Most listeners agree that remarkably convincing spatial perspectives can be achieved with this technique when playback is through headphones. I know that others disagree, but I also find such recordings fairly convincing from a spatial perspective even when played back through loudspeakers if the speakers are placed so that the subtended angle is less than 60 degrees, such as is usually the case when the speakers are set up by the rule of thirds pointing down the long dimension of a rectangular room. You be the judge; here is the site where you can order binaural recordings.
What is needed is a technique which most can agree offers binaural-quality soundfield with loudspeaker playback. I’d like to suggest that the Dimensional Stereo Microphone (DSM) technique being marketed by Sonic Studios, which is arguably a variation of the binaural technique, is just such a system. Truly remarkable results can be achieved when tiny listener-worn omnidirectional condenser mikes (usually mounted on eyeglasses) are played back either through headphones or loudspeakers. If the recording is intended for loudspeaker playback, the microphones are simply moved forward of the ears along the glasses earpieces toward the lenses. See, for example, http://www.sonicstudios.com/dsm.htm and http://www.sonicstudios.com/anarchy.htm. Such recordings can sound amazingly realistic even though the quality of the available mikes for this application is a bit questionable.
One obvious problem with this technique is that the listener’s own personal recordings while wearing the mike probably map better to that listener’s ears than to others' ears. Another set of problems are caused by the self-noise and perspective changes caused by even small movements of the listener/recordist while recording the performance, movements which cannot be compensated for upon playback, even by the same listener who recorded the event.
Great results with this technique can also apparently be accomplished using relatively inexpensive dummy heads as baffles. What you lose to the “averaging” involved with such dummy heads, you gain back in stability of image and repeatability of results. Sample recording files using this technique with a dummy head as a baffle are available at http://www.sonicstudios.com/mp3.htm. These files sound remarkably live even when streamed off the Internet through my cable modem connection.
There are, of course, very few commercial examples of such recordings, this technique having been developed primarily for making "bootleg" recordings of live events from the audience. Many of the products and accessories available from Sonic Studios are aimed at addressing problems created by the need for stealth and ultimate portability of the recording system. These problems would evaporate in the context of serious professional recording equipment. The DSM technique thus strikes me as a potentially profitable avenue for further research since such mike arrays seem to be the best ones available for mapping the recording perspective fairly well to the listening room perspective with two-channel stereo. I would imagine that the high-quality artificial head and shoulders mannequins used for the best binaural recordings would work well with the DSM technique if the mannequins are simply fitted with glasses or other means for mounting the mikes a bit forward of the dummy's ears. Combine this with extremely high quality miniature condenser mikes and we may really have "something to write home about," as they say.
More Channels Needed
But most theorists agree that for truly accurate three-dimensional spatial mapping of sonic event, more than two channels are necessary. The Sonic Studios folks have experimental surround-sound models available where the head is equipped with at least four mikes. See http://www.sonicstudios.com/index.htm and scroll down.
Alternatively, I suggest further research into creating microphone arrays which can move beyond the Blumlein technique's accuracy in capturing spatial perspective within the 90-degree included angle of the mikes. While Blumlein obviously captures sound from the sides, behind, and above, it does not map such sounds accurately onto the soundstage as reproduced by two stereo speakers. This may be the reason why most practitioners place even Blumlein mike arrays closer than the perspective goal would indicate: the room sound is captured from all around the mike array, but with two-channel stereo playback, this extra reverberation comes from the two stereo speakers up front along with the direct sound, adding an extra dollop of concert hall reverberation to the direct sound which tends to make the source sound further away because of the increased ratio of reverberant sound coming from the same general direction as the direct sound of the instruments.
Perhaps all that is needed is for the information captured by a single Blumlein pair to be extracted and distributed among four or more channels around the listening position. If real height information is desired, a second Blumlein pair aimed between floor and ceiling might capture this information. There should be some way of combining the outputs of one or more Blumlein stereo-pairs aimed in different directions to accurately map a full 360-degree soundfield, at least when played back through some reasonable number of channels. I will leave it to the mathematicians and other theorists to figure out how many channels are necessary and how the microphone pairs should be oriented and their outputs summed or subtracted to yield such a result. Maybe the solution is as "simple" as a resurrection and enhancement of the Ambisonics technique. Note that REG used some sort (he hasn't said exactly what he did) of Blumlein-related technique to produce very convincing rear-channel surround on the Water Lily Mahler 5th Symphony SACD recording he was involved with.
I do know that with DVD, SACD, and Blu-ray, we now have commercial media which can give us at least five channels to play with in an attempt to accurately recreate a soundfield. Whether five or seven is enough to unambiguously map the soundfield from recording hall to listening room I do not know, but we now have media which can at least get us closer to the absolute, if that if where we want to go.
I suspect, however, that the vast majority of commercial recordings will never go there. Perhaps I am overly pessimistic about such things, but it seems quite likely that commercial releases on surround-sound-capable media will move in the direction of an attempt to put the listener in the middle of the musicians, and that the limitation of the extra channels to ambient effects will be deemed too subtle and “wasteful” of all those extra channels, especially given the bleed-over in practice and expectations from the established home theater market, where such sonic shenanigans are already the norm.
“They Are Here” as a Parlor Trick
"They are here" is a lot easier goal to attain, but is really on the order of a parlor trick, especially for a single instrument. Record any instrument with good mikes in a single-point stereo array from quite close up and you will effectively ignore or erase the room sound of the recording venue. Play that recording back with flat speakers (well, maybe a bit of a HF rolloff to compensate for the very close mike positioning in the recording), with reasonable dispersion and low distortion, at the right volume level in a relatively live room and--Presto!--you have "they are here" in the listening room, since the listening room acoustics don't fight with recorded acoustics and thus the listening room's sound takes over almost completely.
Acoustic Research once used a twist on this technique to confuse listeners in a blind test in a concert hall as to whether the orchestra was playing or whether the sound was coming from a battery of their speakers set up on stage in front of the musicians. It worked because the recording of that orchestra was made outdoors or otherwise anechoically. When the recording was played back in the hall, the concert hall acoustics so swamped the direct sound from the speakers that listeners supposedly couldn't tell the difference.
As others have noted, "they are here" really doesn't make literal sense for large groups anyway since you could not physically get an orchestra into a listening room.
Listeners who prefer the "they are here" goal usually do so because, in this day of movies and home theater, many crave the overt aural stimulation that immediacy and high volume yield: a surreal or "better than real" sound that "sounds good" to them. I can't say I totally blame them; I like the added immediacy of close seats even for large orchestral groups and actively dislike listening to small acoustic instrumental groups from far back in a hall. There is something to be said for higher volume, more immediacy, and exaggerated staging and envelopment to make up for what you lose in not being able to actually see the musicians before you.
[Concluded in Part Three. . . .]