In following ongoing audio research over my career I'm aware of the many failures the quest for a full-range, active electronic system has engendered. Many have tried and documented their efforts, even in failure. From time to time in other fields, medicine for instance, we've read that a breakthrough on a particular problem has finally taken place because a researcher has looked at a problem from a totally different perspective and it is this totally different perspective that causes the breakthrough.
This is my understanding of how Audyssey came about. Tom Holman, as related in our Audioholics interview, posed the problem, then defined the parameters within which the project would be considered successful. In support of the Audyssey piece I was writing for Audioholics after my interview with Chris and Tom, Chris Kyriakakis sent me the 11 articles which his group have already presented to various peer associations such as IEEE and AES. If Gene has sent you these papers you'll see that Audyssey's multi-faceted mathematical theory was postulated and refined by Sunil Bharitkar, Chris Kyriakakis and Philip Hilmes.
The 11 presentations describe the various innovative approaches these three men took over the five year or so of development it took these to finalize Audyssey which is actually the project name under which a variety of different technologies were perfected. Exactly how and to what degree various turns were taken or decisions made within this basic framework is where "intellectual property" takes over while patents and/or other methods of protection for their work moves forward. This tact is common and intelligent practice for protection of intellectual property.
Next point>>As Tom Holman explains in our interview, THX is a methodology for eliciting the most faithful-to-the-director's-intent recreation of his acoustic vision within the confines of (first) movie theaters and later Home Theaters.
The problem with THX, if you want to label it a problem, is twofold.
a) Until recently, there did not exist a "holistic" program for a home theater to meet all "THX'-intent" criteria because the room acoustics were not taken into consideration.
b) Even now, it is virtually impossible for THX Home Theater certification to be achieved without calibration and sign-off by certified professional installers which means >
c) Cost, as is almost always the case, becomes the main issue.
Audyssey is an attempt to drastically reduce the cost of such performance in a Home theater environment. Given my two limited listening sessions with the product, both of which I reported on for Audioholics, IMHO this technology works exceptionally well and should easily fulfill its promise as a cost effective, legitimate and audible set-up, calibration and room correction solution.
In time of course, the system will be backwards engineered and its inner workings reported on. Buy my point here is that we don't, we can't, in fact know these answers ahead of time. We can only report on results of our listening. I will confess that I've come to a few of my own conclusions just mulling over the facts before us all. But supposition without extended listening is merely guessing. And you know what that's worth.
For now I give the whole Audyssey crew the quiet respect I feel they've earned.
Defining some of the other systems.
R.A.B.O.S. is a product using a 1/20th octave parametric equalizer which has been on the market two years. It works in the 20Hz-100Hz region. Any user who works with this manual system can see the before and after (at the listening position(s)), room curves. My take as to why this system has not been more successful is that most people, even reviewers, are just too darned lazy to take the time to set this system up properly. Once, you do it, and hear the incredible difference this system achieves, you're trained to always listen for this flat, tuneful, realistic bass in every system.
The Lexicon system, just by juggling a couple of numbers that they do publish, would appear to be capable of over 1/300 octave resolution or 0.73dB from 20-250Hz. And it's automatic. So this system would seem to be even more accurate than the R.A.B.O.S. within this specific (and expanded) region. I did hear a short 3 minute demo of this system at CES and it sounded great. The sub/full-range speaker blend the Lexicon/Revel team achieved in their demo room was about as good as I've ever heard. I plan to report more on this system when I interview Dr. James Muller who has graciously already accepted my invitation.
The one active system that I heard that did a pretty good job (still did nothing for RT60, little for nulls, etc.) was the Tact. However, when you consider the cost of something like that, I can get another sub to help smooth things, get a basic analog parametric, fully passively treat the room, and have money left over. I'm not saying it's not a good solution but it's certainly not for everyone.
We bought one of the first Tact 2.0 systems at Alesis back in '96 to use as a loudspeaker development tool. IMHO the Tact was excellent for the purpose I had in mind whith was to chart 9 points in space around a listener's head. This mapping was meant to define the listening window for a mixing engineer using the Point Source D'Apollito studio monitor I was designing at the time. I listened in mono, with the speaker just off the center of in a room with an acoustical tile ceiling. Heavy theater curtains were draped all around the DUT and the mic position.
In the times I attempted to use even a single speaker mic combination without my semi-anechoic atmosphere it immediately became apparent that
a) the room characteristics where intruding on the measurement and
b) that the resultant speaker "equalization" did not sound as good when the theater drapes where again placed around the speaker and listening position.
Tom and Chris made two points during the interview which became more clear to me after I ingested then pondered those 11 papers.
a) To get good sound in several positions requires hundreds of points around each listening position which define a listening window within a room.
b) There is a defined time-window, in this case 200 milliseconds, which will give the best information on direct versus reflected sound.
The last points which I added into my Audioholics Audyssey article are:
a) The room was already passively treated to yield an exact, (down-the-middle of the .2 to .4 millisecond reverb time recommendation window). 0.3 second, Rt60 reverberation time.
b) From my understandning, sound energy, by octave, is re-distributed, to the greatest extent possible, using proprietary techniques. And it is claimed that the signal, at the listening positions, remains minimum phase across the system's entire bandwidth of operation.
Now can we please get back to answering real world questions for our readers who take the time to ask?