Is THX now saying Dipole speakers out for Discrete formats and direct radiating in???

It is not a fact.


The person that posted the information has an irrational persistance that has lead to misinformation.


The THX Ultra (original)Home Theater positon on dipoles has NOT changed. They are still required for surround channels (including EX rears).


THX Processing and design specifications apply to all surround formats for home theater.




------------------

Irv Kelman

We are THX Certified.

Well, for what it is worth, which is nothing more than the time it takes you to read this, I am a believer in direct radiators also. Even when using matrix sound such as Circle Surround and ProLogic, I like direct radiators, too. I know I am probably in the minority, but that's life.


Best regards,




------------------

Oscar Neundorfer

Chief Engineer

SMART Devices, Inc.


[email protected]

Direct radiators have "problems", perhaps better stated as "issues" that must be addressed for proper placement in smaller venues. (And, I'm not about to get into a personal preferences debate here.) Issues which do not exist with dipolar designs. (Among these are the fact a monopole's timbre will change with relative height whilst a dipole's will not.)


I wrote a series of articles on room acoustics that addressed the dipolar approach to soundfield recreation that I'm not going to repost at this time. None-the-less, when one examines the characteristics of sound fields created in concert halls, jazz clubs, churches and theaters these fields are broken into several parts...such as the near field, far field, (even the far out field) and reverberent field. When one examines the physical characteristics of (if not the technical definition of) the reverberent field we'll find that it simply cannot be created (in a small space) with monopolar or direct radiating designs. Hence, the use of dipolar designs for this purpose.


Clearly, there may be mixes (music and otherwise) where the sound engineers were not attempting to create a reverberent field but rather a highly localized sound field to the side or rear. Creating such localization is problematic for dipoles almost to the same degree that monopoles have in creating a reverberent field. This could be a rather sad practice because it would require that (a) the listener know in advance how the sound track was mixed; and, (b) the listener have both dipolar and monopolar speakers and can easily switch between them. If we think back to Quadraphonic, this was a system of four direct radiating speakers. It is often suggested that Quad failed because it was used to create non-realistic effects that soon became tiresome (and, it could not recreate a reverberent field.)


------------------

D. Erskine

DEsign Cinema Privee
www.DEsignCinema.com

Imagine what you could do, if you could do all you imagine.


[This message has been edited by Dennis Erskine (edited August 20, 2000).]

Quote:
This fallacy was addressed in a previous thread at http://www.avsforum.com/ubb/Forum3/HTML/000950.html

and I sincerely hope no further "elaboration" will be forthcoming to perpetuate it.


There is nothing in the Meridian or Lexicon 7.1 algorithms targeted to a specific kind of surround speaker. The processor can neither make assumptions nor adjust its processing for a specific type of speaker, because it only feeds one signal--whether to a monopole or to both sets of drivers in a dipole.


Lexicon is adamantly neutral about the type of speakers to use with their surround processors, and any claim that Lexicon owners are "better off" with dipoles is hogwash. The actual processing in Lexicon's surround algorithms (and even standard THX, AC3, and EX rear matrix extraction) are in no way dependent on the type of surround speakers used. these effects are based on time delay offset for the rears and sophisticated comparison of front channel and surround content to derive appropriate steering; driver polarity does NOT enter into it.


What DOES enter into it are room and speaker placement issues, which remain the same regardless of which type of processor you're using. The pros and cons regarding dipole vs. monopoles are inherent in the speakers themselves and have nothing to do with the software algorithms involved.


The other fallacy in this thread stems from erroneous newsgroup posts claiming THX has "dropped" dipoles. They have not, as Irv correctly pointed out. The "source" cited in the newsgroup post turned out to be a Lucas scoring engineer (not even a soundtrack mixer), and not a THX employee, who was expressing his own opinion about home theater setups, not speaking as a representative of the THX company position. FWIW, Gary Rydstrom, the sound engineer who mixed "Jurassic Park" uses dipoles in his home system.


But personal preferences are irrelevant: the THX position remains a matter of public record: that dipole surrounds yield optimum results, even for 5.1 use. Although THX will now license monopoles under their lesser "Select" standard for small rooms where dipoles will not work properly, even those speakers must meet wider dispersion characteristics than a standard monopole design will accommodate. This is why the M&K THX Select speakers have their tweeters firing away from the listener and towards the rear for a more diffuse effect--a poor man's dipole, which is exactly the niche for THX Select. For the more demanding THX Ultra standard, dipoles remain mandated.



It's worth a reminder that the question posed in this thread asked about the THX position on dipoles. Therefore, basing answers on assumptions drawn from one's own preferences are irrelevant at best and misleading at worst--it is confusing what should be kept a separate issue.


The dipoles vs. monopoles debate remains an ongoing controversy because there really isn't a clear-cut winner. The truth is, each design has its strengths, and the one that works best will depend on the intended system usage. I have heard each perform well in different systems and rooms. You will find respected audio experts advocating either design. Meridian's Bob Stuart believes in monopoles (though again there is nothing in the Meridian surround processing algorithms per se that makes them any more suitable than dipoles; however, Stuart's belief in monopoles is reflected in Meridian's own speaker designs). Aerial's Michael Kelly, on the other hand, is adamant that dipoles make the best surrounds. That puts you in an awkward place, Jerry--unfortunately, there's no spoon-fed formula to be sure of. But here are some of the trade-offs involved:


Dipoles have a more expansive and enveloping sound, and are more effective in creating a sense of ambient space. Thus, environmental sounds like wind, rain, crowd and street noises are typically more convincing with dipole surrounds. And since these kind of sounds make up 99% of the soundtrack in a movie (even 5.1 mixes), dipoles are very effective, especially in a 5.1 speaker setup. However, dipoles have some specific placement requirements--they have to be directly to the sides of the listener, and there has to be space in front and behind them for the wraparound effect to develop. It is also true that you have to be careful choosing dipole designs--many older ones roll off the highs around 12kHz or even lower per THX Pro-Logic specs....in a 5.1 system, full high frequency extension is essential.


Where dipoles don't perform as well as monopoles is in the ability to precisely localize discrete sounds in the surround channels. However, monopoles have limitations of their own for surround use. A monopole speaker by design is inherently exclusively directional--the only way to achieve diffuse effects with monopoles is through phantom imaging between pairs of speakers (i.e., surrounds and fronts)--and that requires them be placed within a certain distance from each other for a phantom image to develop. Exactly the same requirement as creating a frontal 2-channel stereo image--move the speakers too far apart and the soundfield collapses into the individual speakers. If the distance between monopole fronts and monopole surrounds is too wide for phantom imaging to develop (once again, more likely a problem in large rooms), then you get gaps in the soundfield. A rainstorm in which you only hear raindrops falling in the vicinity of the speaker isn't very convincing.


In a lot of anti-dipole arguments you'll see dipoles being dismissed as being totally nondirectional, which is too simplistic. The diffuse effect of dipoles does not eliminate directionality in a discrete surround mix--discrete sounds are still directional, they just don't have the precision placement to the inch that monopoles offer. When you consider the actual content of surround channels instead of trying to argue from your armchair, you find that even the most directional (and infrequent) sounds do not require that degree of specificity in the surround channels to serve their intended dramatic purpose in the film.


The way films are mixed, the only precisely directional effects are in the portion of the soundfield that corresponds to the visual field--the LCR array. And the reasons are simple and straightforward. First, the visual field is where the filmmaker wants your attention focused, and too much discrete activity in the surrounds is going to take your attention away from the screen. Second, sound engineers cannot make the kind of assumptions about the audience seating relative to the surround speakers that would warrant that degree of localization. This is as true in the home as it is in the theater, since a "precise" directional cue from a surround speaker behind you is going to be different from one positioned to the side of you, and the reality is that even home theaters take different approaches to surround speaker placement. For all these reasons, the need for precise localization in the surrounds is not--and cannot be--as strict as it is in the frontal LCR array that corresponds to the visual field. This renders the issue far more open to debate...again, my experience has been that either type of design can work very well in the right setting.

Quote:


The situation only gets more complicated when we move beyond film soundtracks to the multichannel music arena. Here, many discrete 5.1 mixes benefit from using the same kinds of speakers all around, since instruments reproduced at the sides and rears should sound the same as when they're in front. Monopole surrounds are not as prone to the kind of rolloffs and timbre shifts you find in dipoles, and hence serve SOME multichannel music mixes better.


That SOME caveat refers to the fact that there are two different approaches to multichannel music mixes. The first puts you in the middle of the musicians, so you're surrounded by them (the "You are There" mix). The second attempts to create a traditional soundstage, with the instruments in front and the surrounds used for ambient reflections to create a more realistic sense of live performance space (the "They are Here" mix). Obviously, monopoles will better serve the former, since instruments positioned directly in the surrounds will benefit from identical speakers all around. For the latter type of mix, which uses the surrounds primarily for ambience to evoke a sense of space, dipoles can work just as well or better.


To further complicate the issue, EX has changed the rules by requiring an additional pair of surround speakers in the rear. In a 7.1 speaker setup, you're more likely to get an enveloping soundfield without gaps with either speaker design, simply because you're using more speakers. So the issue starts to evaporate...a little. At present, the rear channels are only used in the context of movies, and are not discrete (personally, I don't think the DTS 6.1 format is going anywhere). It remains to be seen whether DVD-A will gain a foothold in the consumer market.


Until then, the best use of a 7.1 setup is not with EX (there are too few EX discs, and regular material sounds horrible through EX), but with the proprietary surround processing in the Lexicon and Meridian pre/pros where you get stereo rear channels intelligently derived from any source. In those setups, you can really hear the future of multichannel music.


A configuration popular with many Lexicon owners is dipole sides/monopole rears. Dipole sides give you a more enveloping wraparound effect and fill in the space between fronts and rears, and the monopole rears give you more discrete localization for hard pans. But again, this can vary depending on the room.



Sorry for the length of this, but the whole issue is quite complex, and the simplistic "[xyz]-type of speakers sound best to me" kind of posts don't serve anyone.


Cheers,

Philip Brandes



[This message has been edited by Philip Brandes (edited 08-21-2000).]

Philip,

Thank you for your detailed explanation. I very much appreciate your insights into the trade offs presented by the various speaker designs.


You seem to assert that I am looking for a "spoon-fed forumula". That is in fact not true. Like most, I am curious about the pros and cons of various designs. What I am looking for is what combination of these tradeoffs works best/ interests me the most. That does not present an ackward situation for me in any sense.


This is actually a situation most of face each and every day: the acquisition of knowledge in order to make a better decision. I am sorry for you if you find that persuit ackward and then trnaspose this ackwardness over to me.


I also asked Dennis about comments he made that one processor is optimized for one way and another for another way. While your explanation makes perfect sense, perhaps Dennis has something to add to this. I dont know as much as he does; and this is why I ask the question.


Philip, again, I appreciate your contribution on this topic I think I learned some very interesting things. Im sure others looking to gain information on this topic will find it useful also.


- Jerry


[This message has been edited by aerialman (edited 08-21-2000).]

Quote:
Jerry,


I'm sorry if my point was unclear.


I do not find the acquisition of knowledge "awkward" and I certainly encourage and try to contribute to it as fully and completely as I can in my posts. The "awkward position" I was referring to was the fact that the respective manufacturers of your speakers and surround processor take diametrically opposed positions on the question of dipole vs. monopole surrounds. This conflict does not lend itself to an easy resolution, regardless of your commitment to knowledge.


And by "spoonfed," I was referring to the general level of either/or positions that accompany the endless debate on this issue, rather than to you in particular. I get tired of reading the same misleading rhetoric over and over, with personal preference and limited experience being trotted out as basis for a universal "right" answer. What I have tried to show here is that the question is far more complex, and that the most important considerations entail factors other than the kind of processor you're using. My caution was that this notion that either Meridian or Lexicon are somehow "optimized" for one kind of surround design is misleading, and diverting the focus from the considerations that really need to be addressed.


I have had my differences with Dennis, but in his post above he is right on the money--each type of design has its strengths and limitations, and he did a good job of summarizing them. I was trying to expand on his summary.


BTW, was your "trnaspose" a typo or deliberate? If a typo--it was a funny one!

Quote:
Bob -


There is nothing wrong with expressing an opinion when it helps resolve a topic under discussion. As a matter of fact, I happen to agree with you that Dynaudios kick B&W butt--and would happy to discuss the reasons in an appropriate thread.


However, THIS thread, as the title clearly shows, is not about personal preferences re: dipoles vs. monopoles--it is about the THX POSITION on surround speakers. The latter is not a matter of opinion, but of fact. And as the initial post here shows there is some confusion about what those facts are.


That is the distinction I was trying to make.


Cheers,

Philip Brandes



[This message has been edited by Philip Brandes (edited 08-21-2000).]

Philip,

I got a good laugh there- yes that was a typo and a totally innocent one!


I am confused by these statements, you just said:

Quote:
and earlier you said:
Quote:
could you elaborate on this? I agree with your quote directly above. As to Aerials dipolar surrounds, I need to point out I have no affiliation with Aerial Acoustics, Im just a fan, but I remember reading over at SMR that they are coming out with a series of direct radiating surround speakers based on the great and popular CC3 center channel. These arent to replace the dipolar SR3s, but to add a direct radiating surround speaker.


now, I have a friend whose opinion I respect very much who thinks the way to go is direct radiator all around.


my opinion is that maybe the "compromise" position is best- do dipolars on the sides and directs in the rear- since my room will be layed out as a rectangle and done so length wise- the dipoles could help bridge the space between front and rear- this is just my uneducated guestimate on this. I was also thinking maybe of using a good center like the CC3 for the sides as the CC3 has a pretty wide dispersion pattern.


thoughts?


- Jerry

Quote:
I'm not sure what else there is to say. There is no conflict between these two quotes. One is talking about designer's preferences for one type of surround speaker over another, but that is based on the inherent properties of the speakers themselves, not the surround processors. The second is simply addressing the fact that a processor either decodes or synthesizes a single signal to each surround speaker, regardless of whether it using a monopole or dipole driver array. There is nothing the processor can do to adjust or compensate for the type of speakers you're using--the difference in sound will be totally a function of the speakers themselves. And even if such tweaking were possible, which it is not, it would be economic suicide for Meridian to bias their system against the official THX stated specifications--at a minimum it would cost them THX certification.


As for the divergent opinions, Bob Stuart obviously believes monopoles make better surround speakers, since his Meridian surrounds are monopoles. Michael Kelly has stated many times that he thinks dipoles make the best surrounds even for 5.1 recordings. He told me this personally at CES last Jan. and will say so to anyone who calls and asks, as others have posted in these forums.

Quote:
Sure he is, because customers are requesting it, and because the specific placement requirements for dipoles rule them out in some installations. You can see a picture of the LR3 in my CES show report at http://www.smr-home-theatre.org/ces2...kers/index.htm


Nevertheless, Mr. Kelly is quite firm in his own personal conviction that dipoles make the best surrounds.

Quote:
I have friends and industry sources on both sides of the question.

Quote:
It sounds like a dipole side/monopole rear combo would work very well in your setup, especially with the long walls at the sides. As I said above, this is a popular combination among owners with 7.1 speaker setups. The SR-3s offer the additional flexibility of being configurable as bipoles, so if you don't like the dipole null effect you can set them for a more directional sound. If you're doing head-to-head comparison of the SR-3s in both configurations, however, remember that the dipole null cancellation results in an overall 3 dB drop in perceived volume, so you need to recalibrate your speaker levels each time you switch in order to make a meaningful comparison. It is the lack of awareness of this important difference that has contributed to dipoles' reputation of being "wimpier" than monopoles.


Cheers,

Philip Brandes

Quote:


Philip,


I understand most of the issues involved with the various speakers (as you so clearly overviewed). But the one thing I am having a hard time understanding is the frequency extension of dipoles. For example, I see M&K quoting a 20kHz extension on their S-150P, yet the Surround-250 (which has an identical driver array) is shown on p.52 in the Fall issue of Home Theater Buyers Guide to taper off quite a bit after 10 kHz. In the monopole mode the speaker extends to 20 kHz.


I am just using this speaker as an example, but I am wondering if most dipoles have limited performance past 12 kHz, even if the radiators on each side may have extension to 20 kHz? I read somewhere that the higher frequencies get focused, kind of like laser beams, and therefore do not get reflected off the immediate surroundings and make it to you're ear. If this is the case, that could be remedied by using a smaller aperture, for less focusing. Or is that the higher frequencies attenuate more on reflection? I really don't know. I'm just having a hard time understanding why on every review I see of dipoles, their is a significant dip in amplitude past 12kHz? Is it really a Pro-Logic legacy, or is there

something inherently difficult (perhaps tiny apertures required) in making

a dipole with full extension?


FYI, I have a Lexicon MC-1, with 5.1 direct speakers, and I am planning on adding dipoles to the sides, since I tested, and enjoyed the diffuse sound. I'm just trying to understand which brand of dipole I should ultimately get.


Dan

Quote:
This is a horrible, but not uncommon position to find yourself in. Because of the way we humans perceive, or localize, sound it is important to make certain your surround speakers (indeed all your speakers) are timbre matched. This can be a challenge within a manufacturer's line and almost insurmountable across manufacturer's.


------------------

D. Erskine

DEsign Cinema Privee
www.DEsignCinema.com

Imagine what you could do, if you could do all you imagine.

Quote:
Can a surround processor take advantage of a dipole's polarity for steering of sound when four surrounds are in use? The answer to this question is yes and it has been done. Philip is correct that it absolutely cannot be done in any mode on the processor denoted as "THX". I might also suggest that to take advantage of such characteristics, speaker placement becomes tremendously difficult.


When it comes to Meridian processors and their monopolar design digital speakers, I do have to suggest that very significant benefits can be achieved when the processor can communicate with the DSP processors in each individual speaker --- particularily when one can take advantage of Bob Stewart's very significant expertise in psychoacoustics.



------------------

D. Erskine

DEsign Cinema Privee
www.DEsignCinema.com

Imagine what you could do, if you could do all you imagine.


[This message has been edited by Dennis Erskine (edited 08-22-2000).]

Philip,

my confusion is based on three parts:

1- On one hand you say surround processors themselves, such as meridian, cant favor one type of speaker design over another;


2- Then you seem to be implyinig that since meridian the speaker manufacturer favors direct radiating speakers. Therefore meridian the surround processor must work best with direct radiators.


3- Aerial the speaker manufacturer seems to favor dipoles for surround sound.


So my confusion comes from parts 1 and 2 and this statement by you:

Quote:
are you saying Meridian the surround processor company and Aerial Acoustics are in conflict with each other over what kind of speakers to use as surrounds? This is what is confusing to me. I realize there is one Meridian. Youve said in the past that the surround processor just passes the signal along to the speaker. It seems to me what you mean is that Messers Stewart and Kelly's approachs to surround sound when making speakers seem to be different- one preferring direct radiating speakers (Meridian) and the other dipolar (Aerial).


I now have a new area of confusion:
Quote:
Are you speaking for Mr. Kelly? "personal conviction" are strong words.


While were on the topic, Dennis, I hope you wont mind my borrowing your quote because I think that a truly horrible position is that you dont have enough food or shelter- our problems (dipoles vs. monopoles) are the problems I wish we all had as our biggest problems. [end of soapbox]


- Jerry



[This message has been edited by aerialman (edited 08-22-2000).]

Quote:
Dan,


The reflected nature of dipoles does result in some high-frequency loss. In better dipole designs (such as Aerial, Snell, B&W), you'll find the tweeters are biased to play hotter in order to compensate. Even so, you'll find that dipoles do deviate from a "flat" response in more ways than just HF extension (room acoustics play a larger part than they do with monopoles), although high frequency energy is still sufficiently present for the audible spectrum. Older Pro-Logic-style dipoles rolled off at 8 kHz, which is completely unacceptable for for 5.1.


On a related issue, the notion that the dipole null eliminates direct energy from the drivers is also a misconception. What it does is reduce the level of direct energy by about 3 dB--in other words, you are changing the ratio of direct/reflected sound, not making it all reflected.


BTW, on a separate note, one potential source of confusion in your MC-1 is the "Surround Rolloff" parameter in some of the effects modes. This rolloff is ONLY applied to additional signal generated by the MC-1--any sounds present on the recording are passed through regardless of where you set the rolloff. The parameter is only there to let you adjust the amount of "liveness" in the MC-1's added processing. (and it has *nothing * to do with whether you're using dipoles or monopoles : )


Cheers,

Philip Brandes



[This message has been edited by Philip Brandes (edited 08-22-2000).]

Quote:
What I am saying is that when it comes to what type of speakers to use for surrounds, Bob Stuart and Michael Kelly have diametrically opposed views on what designs are optimal. This is exactly what I'm saying and this is ALL that I am saying. I neither said nor implied that just because Bob Stuart prefers monopoles that his surround processor *must* work best with monopoles, and I fail to see how you could be reading this into my posts. I am only saying that he designs his SPEAKERS according to his preference. His PROCESSOR is designed to support any type of speaker connected to it (as are ALL processors). I really don't know how to say this any more clearly.


Quote:
Yes, as I stated earlier in this thread, this was what he personally told me at CES last January. Unless you are accusing either him or me of lying, I would have to say it represents his position on the subject.


In addition, there have been many first-hand accounts of his advice posted on various sites. Here is one from Digital Theater Forum dated Feb. 1, 2000:


"After I got all the deals done, I decided call Aerial and see if they

had some stands available for my new 7s. Didn't really expect anyone to

answer, but they did. Spent some time talking, and found out I was

speaking to Michael Kelly, the president of Aerial. He must have spent

20 minutes on the phone asking about my setup, how it would be

configured, etc. (He was a little skeptical about using 5s for the

rears, as he personally does not like direct radiators for his surround

speakers.) I asked about the performance of the 7s vs. the newer 7b and

model 6. He was very generous with his time and quite enjoyable to speak to."


That last observation is one I agree with. Call him yourself if you have any doubts.


But this is not to say that Mike Kelly's--or Bob Stuart's--personal preferences for surround speakers is definitive. It is simply pointing out that there is room for disagreement on this issue even among well-respected industry professionals, and that's because neither dipoles nor monopoles offer an entirely perfect solution.


Cheers,

Philip Brandes




[This message has been edited by Philip Brandes (edited 08-22-2000).]