Albert Von Schweikert of Von Schweikert Research is a highly respected loudspeaker designer who has served as a design consultant to several other loudspeaker companies in recent years. Now at the helm of his own high-end company, his new loudspeaker line is receiving widespread critics praise. Editor Gary Reber interviews Albert Von Schweikert for this exclusive On Screen feature.

Gary Reber, Widescreen Review: One of the major things that separates the many speaker designs on the market today is radiation pattern. What are your views on point source, line source, bipole, and dipole designs?

Albert Von Schweikert: Well, first of all, Gary, I did several years of research to define the target goal for my loudspeaker design. That goal was to simulate the sound of live music, whether for stereo music playback or film soundtracks. Essentially, since most all sound waves that occur in nature are spherically radiated and are not directional, my target was for the speaker system to radiate in similar fashion—that is, a point source radiator without directional beaming. Interestingly enough, this also describes the typical microphone pickup pattern. This is relevant because recordings can be regarded as an encoded source, so I have designed my speakers to behave as the decoder. I call this technique Acoustic Inverse Replication.

WSR Reber: Please expand your views on spherical radiation.

Von Schweikert: A spherical source is more equivalent to a point source than a line source, dipole, or bipole. Many designers forget the fact that a speaker should sound like a live instrument—so it should behave in many respects like an instrument, or more importantly, like the recording microphone in reverse. Many designers are overly concerned about room effects and thus try to get rid of these boundary reflections using the directional qualities of line sources and dipoles. The problem is the radiation patterns of dipoles and line sources do not exactly duplicate those of most instruments or mics.

Might this be a problem? Let’s put this into perspective: It should be true that the goal is to reproduce the original recording of the instrument, voice, or sound effect, right there in front of you, between your speakers. That original sound wave would have radiated spherically from that location into your room, involving all of the boundaries, including the floor, walls, and ceiling. Your ear/brain hearing mechanism can tell how the sound radiated from the original instrument, and can determine if the loudspeaker system is radiating the sound in the same manner. If the radiation patterns don’t match, you can easily tell you are listening to "canned" sound. Localization of sonic patterns is determined by psychoacoustic cues such as timing, phase, and frequency response differences heard at each ear, and are processed by your brain to determine thousands of data differentials which are then stored in the memory for later recall. I have conducted several years of experimentation to prove my theory, and several leading researchers agree with my thesis completely.

The whole idea of trying to eliminate the entire sound of the listening room does not make sense, that’s almost like trying to play music in an anechoic chamber, which would sound very unnatural. Try listening to music outdoors, it sounds quite "dead". The truth is, humans grow up listening to natural sounds in rooms, whether it is parents’ voices, pots and pans dropping, or whatever. We are conditioned to expect a natural reverberation when listening in a room, so I believe that loudspeaker systems can only be successfully designed by taking the room’s effects into consideration and making these effects part of the total design.

WSR Reber: Yes, I actually agree with that concept. Many speaker designers who subscribe to this "keep the room out of the equation" maintain that in the recording itself, through the microphone, the room environment of that original recording is captured, and therefore, it should be reproduced by the speakers alone without room interaction in the playback environment. How do you counter that argument?

Von Schweikert: Of course, I understand that. However, that holds true only if the playback room is highly reverberant, but whose is? Most rooms are very well damped by rugs, stuffed furnishings, drapes and acoustically treated ceilings. In addition, we become used to the room’s reverberation pattern and learn to "tune it out" while listening, just like in the old days of vinyl surface noise. Your ear/brain hearing mechanism is very smart!

Also, the fact that original recordings have a large amount of acoustic or electronic reverb in order to make them sound "better" simply swamps most room effects. Have you heard those extremely dry Denon recordings that were done in a semianechoic environment in order to test your room’s acoustics? No room I ever heard those recordings in could add enough reverb to make them enjoyable!

Another example. Let’s say you are listening to a pipe organ recorded in a cathedral and you hear the immense hall reverberation superimposed on top of your room’s slight reverb pattern. Your room’s reverb is not going to be very noticeable, if at all, once you have familiarized yourself with the given conditions. Maybe you have noticed that it takes a while to get used to someone else’s sound system. I believe the conditioning time is around five minutes or so in trained listeners, and quite a while longer with neophytes. The average listening room’s main contribution is bass reinforcement due to boundary effects, which in most cases helps deep bass response. So even this is not a great problem. Try listening to speakers outdoors if you want to know how bass sounds without a room. I’ve played in bands at outdoor concerts, and the sound was quite unsatisfying!

WSR Reber: So basically what you are saying is that it is not necessary to attempt to subtract the listening room out of the playback system.

Von Schweikert: I don’t believe so. I’ve listened to recordings played back in anechoic chambers and they don’t sound good. They sound dead, as you would expect! We once constructed a live end dead end listening room, which was very much in favor in the 80’s, and after two weeks, we dismantled the room. It simply sounded highly unnatural. Your ear/brain hearing mechanism simply is listening for the cues that it hears everyday in that room. When those reverberation cues are missing, you take notice. Remember if a live instrument plays in your room, you will hear the room’s sonic signature superimposed upon the sound, for better or worse, so why should loudspeakers behave any differently? In fact, unless they have the same spherical, semi-omni-directional radiation pattern that a guitar has, for instance, then the speakers won’t sound precisely the way the guitar sounded, in that room.

Remember that I am taking for granted that we are talking about average, good sounding listening rooms. If someone had the misfortune to live with glass windows all the way around, a hardwood floor, and a hard plaster ceiling that’s a different story! In this case, the room must be treated, or a new room chosen for listening. We definitely don’t want to listen in an echo chamber either!

WSR Reber: The next question I have for you is how these different designs interface with the room also differentiatiates them. Are your speakers designed to be used at or near the room boundaries, away from the room boundaries, on the floor or stand mounted?

Von Schweikert: All of our main speakers are floor standing. I believe in wide frequency response, so we have designed bass response to extend down to the 20Hz range with the larger models such as the VR4 and the VR3000. I don’t believe that many people are willing to alter their living space to accommodate large speakers placed out into their room, so I have designed our new line of Virtual Reality Theatre speakers to be used within 3 inches of the walls. Also since it is extremely difficult to eliminate bass lift due to boundary reinforcement, I actually use it to benefit bass extension and sensitivity. Years ago Roy Allison did work along these lines, which still holds true today.

WSR Reber: You said you believed in a floor standing design. Why not stand mounting?

Von Schweikert: If you put a speaker on a stand, you are wasting floor space that could have been used as extra air volume to extend the woofer to lower frequencies. I don’t believe many people have problems with floor standing designs as long as they don’t take up much floor space. Our new line of Virtual Reality Theatre narrow profile speakers are only 7.5 inches wide and are no larger than a mini speaker mounted on a stand. However, how many minis go down to the 20Hz range? We have chosen the height to insure that the midrange/tweeter array is at ear level in regards to seated listeners. Although many mini speakers are known for excellent imaging due to small baffle width, we are in essence achieving the same thing due to our narrow profile, but our subwoofers are built into what would have been the side of the stand. This is space efficient design, which makes sense in all aspects, from both engineering and cosmetic standpoints. Very few Significant others want unsightly subwoofer "coffee tables" hanging around the room in today’s smaller living spaces, especially since they don’t look very "chic!"

WSR Reber: Should speakers used in a multi-channel array be voiced differently than stereo speakers, for example, flat vs. a warm tonal balance?

Von Schweikert: Well, this brings up the age-old question: if recordings are bright, should the speakers compensate? If I going to live in the Ivory Tower and speak from a theoretical standpoint, the recording engineer should not add sound effects such as treble boost, so that us speaker engineers could design perfectly flat speakers. However, since this is not the case, and we all have heard both music and film sources sound horribly bright, I believe that either the playback electronics or the speakers should have some type of compensation to adjust for excessive brightness. Our top two models of Virtual Reality Theatre speakers, the VR2OOO and VR3OOO, have rear mounted "Spatial Dimension" controls which adjust the level of the rear ambience drivers. In effect, this control not only affects the amount of depth and soundstage width, but also adjusts for excessively bright program sources by reducing the amount of treble reflected off the rear wall. In addition, our center channel speakers have a unique timbre matching control called "vocal Presence." There is a side-mounted potentiometer, which allows the voices to be brought up or down in the mix. I have found that some soundtracks have very high levels of mid treble boost on voices to "blow" through the ½ screen at theatres, and this boost is not pleasant on accurate home equipment that lack controls on the preamps or processors. Our exclusive Vocal Presence control allows for both equalization of voices or timbre matching with speakers other than our own.

WSR Reber: So you have an accurate flat center position, with a boost or cut at the extremes

Von Schweikert: Right, exactly. From a theoretical standpoint, I want to make the speakers flat, but due to bad recordings, we have to add the adjustments. The great thing is that you can always get back to flat

WSR Reber: Well, in motion picture sound tracks, there is a reason why the high frequencies sound boosted in a small home environment, but most processors today incorporate either a frequency roll off curve specified by the Home THX™ Program or another type of frequency roll off curve that is very similar. When musi6 is played through that processor, you can bypass that and just go flat.

Von Schweikert: Right. I think that this is a step in the right direction. As far as the voicing of speakers, your question could also refer to multiple speaker interactions in a five-channel array. Since I believe in flat power response into the room for stereo, the same must hold true for multi-channel sound. However; many people profess to not like flat (read "accurate") speakers. This is due in part to any midrange and treble drivers, which have distortion, coloration, or "glare," and also to inexpensive electronics that do not have smooth highs. Driver distortion is caused by either a non linear motor or a flexing diaphragm. Many polypropylene midrange drivers have cone flex distortion up to 34 percent, while soft dome fabric tweeters have even more, especially at high volume levels.

I have developed cone materials made from woven carbon fiber thread which almost eliminates the diaphragm flexing distortion, while the aluminum dome tweeters I employ do the same thing. In addition, I do not allow the drivers to go into nonlinear operation by using slow crossover slopes such as first order. Instead, I have developed new crossover topologies I call Global Axis Integration Networks, which have many benefits. First, the acoustic slopes are very steep, approaching 24dB per octave. This eliminates coloration caused by vocals being reproduced by woofers and bass being fed into tweeters, both famous problems of first order crossover designs. My GAIN design uses no more part in series with the drivers than many first order circuits, since I use shunt circuits in the form of Zobel networks. This permits very high transparency, yet allows little driver overlap, the cause of both driver distortion and beaming problems called "lobing." The Global Axis Integration design allows the speakers to radiate very wide dispersion with better than average depth due to the rear firing ambience drivers, and there is almost a total absence of beaming or lobing effects. You can stand then sit, or move several feet off the horizontal axis and hear little change!

I have found that off axis frequency response is as important as the axial response, since you will be able to hear problems due to reflections from the room boundaries. Many engineers do not realize that first order crossovers are only phase coherent on a very small window directly in front of the tweeter. Once you put the measuring mic 90 degrees off axis, especially vertically, the signal no longer sums to unity, and there can be up to 12dB dips and peaks in the frequency response, with corresponding distortion of the phase response.

The Global Axis Integration Network allows phase consistency over a very wide polar arc, both vertically and horizontally. Listeners are constantly amazed when they find they don't have to sit in one small sweet spot. They can get up and walk around and still hear correct imaging and timbre! This is critically important for home theatre use since with conventional speakers, only one person will be in that "sweet spot," which is very frustrating for the rest of your family and friends! Everyone can appreciate the benefits of the Global Axis Integration design after they experience their first movie soundtrack through our VR speaker system.

WSR Reber: Do you subscribe to timbre matching of the rear speakers, the idea that the rear speaker should be voiced differently than it's opposite in the front, as THX™ states, to counteract the different head related transfer functions the listener experiences when sounds arrive from the sides and rear? That's their position, their argument. Do you subscribe to that timbre matching argument?

Von Schweikert: Yes, I do, but only under the condition that we are listening to a monaural rear signal, the old Dolby ProLogic method. If you have side surround loudspeakers aimed directly at your head, you can tell where the speakers are placed when listening to a mono signal. It sounds like you have headphones on, and the ambience effects sound like they are originating in your head. Let's say you are watching an Indiana Jones film and he is standing in the rain forest. Naturally you do not want to hear where the rain is coming from, you want the rain to sound like it's all over the room, completely surrounding you. So, under the given situation where we are listening to a Dolby ProLogic soundtrack, even with THX™ enhancement, the rear signal is monaural and we would like to spread the sound around the sides and rear of the room. In theatre applications, this consists of at least four speakers on each side wall and four behind. Since most people cannot install this many speakers, the THX™ designed dipole null certainly is useful, since the room can be "washed" with only one side mounted pair of small speakers. The null caused by the dipole design does not allow the listeners to "hear" where the speakers are mounted, and indeed generates an all-encompassing ambience effect. Now, 180 degrees out of phase with what I just described, I would not use a THX™ dipole system with the new 5.1 discrete digital systems, whether Dolby Digital (AC-3) or DTS Digital Surround. Since the stereo signal presented to the rear speakers must be localized in order to gain the advantages of stereo a null effect would be counterproductive and ruin the imaging. The new 5.1 technology will allow rear sounds to be panned from side to side, like trucks moving across the stage directly behind, and even allow sound to approach you from further back, like someone sneaking up behind you.

This apparent dichotomy gives rise to the fact that two completely different types of rear surround speakers are necessary if you wish to play back both older ProLogic movies (15,000 titles and counting) and the newer discrete formats (less than 150 titles as of this date). I have designed such a speaker and will be introducing the Virtual Reality Theatre Model TS-3 at the Consumer Electronics Show in January of 1997. This all format compatible surround speaker will work in any system since it has switchable radiation patterns and built in subwoofer!

WSR Reber: During the matrix era, for many years, Jim Fosgate designed stereo matrix systems for the rear channels. The most recent version of his technology is called Six Axis, taking the 4-2-4 matrix encoded signal and rearranging the phase characteristics so that there is rear and side wall stereo imaging. This advanced matrix steering also works well on stereo-only material such as music. The signal manipulation is absolutely transparent.

Von Schweikert: I can certainly appreciate Mr. Fosgate's technology since the original soundtracks left a lot to be desired. I am basically a fan of this type of processing as long as it's transparent and does not hamper one's enjoyment of the film. I am diametrically opposed to the stance of the hardcore purist who wishes to hear exactly what's on the recording, even if it's bad and extremely difficult to listen to.

WSR Reber: Going back to the previous question when I asked you about timbre matching, I think you talked more in terms of different loudspeaker designs for the rear speakers. I was thinking more of the "timbre matching" electronic circuitry that THX™ employs.

Von Schweikert: Well, let's put it this way I'm for whatever technology may "fool" you into thinking you are at the scene of the movie, whether it's electronic processing or simply better speaker design. If you're watching a movie and a jet flies over your head, you want that jet to sound the same way in the rear as it did in the front in order for the overhead "pan" to sound realistic. You must take into account the Doppler Effect, which is pitch change due to movement. The classic example is the apparent change in the train whistle as it goes by But what I'm talking about is the apparent gross shift in timbre which results from using full range speakers up front and tiny speakers in the rear. As the jet starts out in the front speakers, the engines sound really powerful and deep, but as the jet flies overhead and the pan switches to the rear speakers, the jet suddenly sounds like a model airplane due to lack of mid-bass energy in the rear. I don't believe that electronic EQ in the decoder is as good of an answer as better surround speakers. I have spent several years developing our switchable rear speakers so that they have a very similar sound to front speakers, especially in the bass/lower midrange areas. I use the jet fly-over to judge the accuracy of the transition, but frequency response measurements will show what's going on, especially polar axis response measurements like I use.

WSR Reber: Would you say ideally that they should be matched to the front? In an optimum sense?

Von Schweikert: Absolutely, as far as timbre matching is concerned. However, I believe that the radiation patterns must be switchable, unless the user is going to use only one surround format, for the reasons discussed previously. As far as the head related transfer function question, I have an experiment for your readers to perform. Rotate in place while listening to a monaural sound source such as a person talking. There is naturally a shift in time, phase, and frequency response cues, and you will certainly hear this as you rotate. However, there is an inherent danger in attempting to design a transfer function into a loudspeaker system to compensate for the fact that there is sound originating behind you. In real life, these time and phase effects from rear sounds enable you to "home in" on their location, so wouldn't it make sense to leave the speakers alone and let your ear/brain hearing mechanism do its job? Normally timbre-matched rear speakers would convey the correct soundtrack information that would sound like it's coming from behind you anyway, if the recording engineer has done his job properly. Since we can't second guess what they're recording back there anyway, just timbre match to the fronts and let 'er rip. Does that make sense?

WSR Reber: It makes sense to me, I'll subscribe to that forever. So what you are really saying is that in an optimal sense, to maximize the experience, the speakers should all be identically matched.

Von Schweikert: I would have to say yes. But that's with digital stereophonic sound tracks in the rear channels, the new 5.1 formats.

WSR Reber: Right. Now, also about your comment about the fact that as this jet air-plane flies over from the front to the back, with limited bass range, even in the upper bass range, it is not going to sound the same as it flies through the back direction. Yet that counters the views of people who believe that bass is non directional.

Von Schweikert: Now, we are also supposing that we have bass management in operation, with subwoofers in use along with the five mains. The problem comes in when you look at 99 percent of the surround speakers on the market and find that they use 3-inch, 4-inch, and 5-inch drivers back there for bass and midrange. It's not a question of bass being directional, it's a question of much shorter wavelengths which are directional, and they're the lower midrange frequencies which subs are not going to fill in. The typical small surround speaker is weak in the 80Hz to300Hz range due to small drivers with small magnets placed in very small enclosures. There's that trade-off made for cosmetics, you know, since people don't generally want really large boxes mounted to their side walls and then there's the problem of unrealistic price points, with mass market pulp magazines advertising junk for under $100 each. No wonder the general public doesn't know much about surround sound done right.

WSR Reber: Well Albert, how do you address that real world scenario? I've always written in the magazine from the point of view of instructing how to achieve an optimum experience. Therefore, if you understand how to setup an optimum experience, you then know when you are compromising. Most people are going to have to compromise, even myself, I have to compromise because not every room can support what I would like to do optimally. You have to compromise somewhere, so how do you approach compromise in that sense where people can support the same identical size speakers, or maybe they have a budget of $5,000 to spend. You can spend most of the money on your two front left and right speakers and compromise on the rears. Know what I'm saying?

Von Schweikert: Exactly, since most people might believe that there might be some consolation from the fact that the rear speakers are not played quite as loudly as the fronts.

WSR Reber: Not in all cases, though. I review over 500 motion pictures a year and I have reviewed every discrete digital release, and some very powerful matrix releases and I can tell you the bass energy is enormous in the back channels, often below 25Hz.

Von Schweikert: I saw a movie called Backdraft, which contained energy down to20Hz in the rear channels. It was an awesome experience, so I started designing rear speakers with response down into the 20Hz range. I believe the rear surrounds to be very important to the overall effect, so I have designed several surrounds with deep bass that are fairly compact. Our TS-2 is a switchable radiation pattern wall hanging unit that has deep bass and very powerful sound projection. However, our TS-3 is the killer; with a 10-inch subwoofer that responds into the 20Hz range. The cosmetic design is what I call a "wall hugger." Although a floor standing speaker; it is so thin that you simply place it against the wall and it disappears. This system has the same driver complement as our front mains called the VR2OOO, including the built in 10-inchsubs, so the timbre matching is perfect. The real kicker; though, is the side firing ambiance drivers that are switchable to dipolar operation if you are playing a Dolby ProLogic film! And this thing is only 7-inches thick!

WSR Reber: So the key word here is compact?

Von Schweikert: Well, let's put it this way, it's compact relatively speaking. Although it's fairly wide (16-inch) and tall (42-inch), it's shallow enough (7-inch) to place against the wall without taking up a lot of floor space. And it's available in both black and off-white.

WSR Reber: Do you favor acoustic room treatment? If so, what type?

Von Schweikert: I do in a limited way. First of all, if the room is normally furnished with an equal ratio of absorptive surfaces to reflective surfaces, there will be a balanced reverberation characteristic already. However, if the sides of the room are not symmetrical, such as glass on one side and bookshelves on the other; there may be a mismatch in the reflected sound which might throw off imaging a bit. Some treatment would be in order on the highly reflective side. However, I caution against totally deadening the room since it won't sound natural. Everyday furnishings such as drapes, couches and/or theatre seating, and carpeting will put things in order if the room is totally empty and too reverberant.

WSR Reber: How about the use of electronic equalization?

Von Schweikert: Well, having been a purist for many years in high-end audio, I didn't even want tone controls on my preamp since I thought they would degrade the purity of sound. However, that was due to the poor quality of tone controls in those days. Today, digital equalization, hopefully transparent sonically is right around the corner which should offer not only frequency response alteration but may also offer room correction of bass standing waves and other anomalies best handled by computer chips.

WSR Reber: Do you believe that what's right for stereo is right for multi-channel in the context of the effect of speaker directivity on the precision of imaging when more than two channels are present?

Von Schweikert: Yes. It is correct to surmise that if the speakers all have response similar to the live source, then using five in correct placement would enable the five channels to decode properly at least as correct as the recording engineers expect! As far as over-lap and interaction are concerned, I have done experiments regarding how far apart the main speakers can be spread, and how to place the center channel in relationship. The front three have to be placed in an arc, so that the center channel is no closer to the listeners than the mains. However, on some processors this can be done by digital delay. Placing the mains too close together, directly next to the screen and sitting really far away converts the sound to mono unless your screen is really wide.

The other part of the equation is speaker directivity. What seems logical at first is to engineer a highly directional pattern into the LCR mains, so that there is little interaction that would blend into mono. However, this directional pattern does not sound realistic, since if you move your head several inches, you are going out of the field of one speaker and into the field of the next. This is a totally unnatural situation and is subconsciously very bothersome. Just imagine going to a concert and moving around there are no gross changes in timbre or imaging. If the speakers do not have a natural radiation pattern, your ear/brain hearing mechanism knows the difference within seconds or minutes and let's you know by increasing your boredom level. This is one of the reasons why certain audiophiles keep changing speakers every few months! They are seeking a sound that is not subconsciously irritating them on some level or other. Of course, there are multitudes of other distortions present, but the radiation pat-tern is one area that is neglected by many designers. They seem fixated on axial response flatness only

WSR Reber: So in essence, again it goes back to the optimal setup dictating identically matched loudspeakers and directional patterns, phase and time consistencies?

Von Schweikert: Right, exactly I am not a fan of speakers that are so directional that if you move your head a few inches you will hear tremendous shifts in the imaging patterns, since this does not occur in real life.

WSR Reber: There is a lot of discussion, Albert, about bass loading methods. Do you favor second order sealed boxes, fourth order vented or passive radiator designs, band-pass, bipole, dipole or other methods?

Von Schweikert: Well Gary, I've designed speakers with all of the bass loading methods over a 20 year period, and I believe that it's the implementation that makes or breaks the design.

Originally, back in the '70's when we were designing our first speaker at Cal Tech, we thought that a sealed system would have the best sound and fastest transient response. We heard a lot about the phase shift problems in vented designs caused by the delay from the vent that could cause "boom"-the hollow internal cavity resonates, acting similarly to a capacitor by storing energy and releasing it later, causing that delay which results in "boom." However, the phase shift and delay can be controlled, as I have discovered, by electrical and mechanical damping. Most designers go for maximum efficiency rather than fastest transient response, and the Theile/Small equations clearly show the response trade-offs. However; you need to custom fabricate the woofer to highly specialized engineering parameters if you are going to have a successful design. We stuff the enclosures with 100 percent Dacron fill to lower the "Q" of the system, which eliminates the hollow resonant tone that audiophiles hate. However; we still have the efficiency gain of venting the rear wave, and we can tune to lower frequencies than the same sized sealed box. For home theatre applications, all of these benefits are important.

Our systems sound as fast and clean as sealed boxes, go as low as transmission lines, and are compact. Our slim profile main towers each have built in side firing subwoofers.

WSR Reber: You mentioned the "Q" factor. How would you define that for our readers?

Von Schweikert: Well, "Q" stands for quality factor and has to do with resonance magnification. If you listen to a woofer playing without an enclosure, you notice that there is no bass. The loss of bass is caused by the front and rear waves canceling due to being 180 degrees out of phase with each other. Once you install that woofer in a box, the front and rear waves can not cancel since they are now separated. However, the air pressure inside acts as a "spring" that the woofer has to push against. This pressure causes the resonant frequency to go up if the box is small, but has no effect if the box is very large. This resonance change caused by the air pressure in the small enclosure can be measured as a rise in impedance. The height of the impedance peak determines the "Q" of the system, and also indicates how deep the bass can go. As the box is made smaller; the "Q" rises, along with the resonant frequency of the woofer in that box. Small boxes have powerful bass at upper bass frequencies, but lose deep bass due to the pressure of the air spring. As the box size is increased, the "Q" or resonance magnification, goes down, and the mid bass range is not as pronounced. As the resonance frequency goes down, the transient response becomes faster; exhibiting less hangover; since the air spring is not wildly affecting the woofer suspension. The trade-off is that the mid bass efficiency weakens, even though the deep bass is now extended lower in frequency These interactions and their trade-offs are called Hoffmans Iron Law, after Tony Hoffman of KLH, described them, circa1960. This was before the time of A.N. Theileor Richard Small, who came a decade later.

When the box is vented, becoming a Helm-holz resonator; the vent tube is usually tuned to a critical frequency below the sealed box's cutoff frequency. This allows the box/woofer combination to go deeper as well as increasing the efficiency, since the rear wave is now put into use. However, as mentioned previously, the hollow cavity resonance can cause coloration, unless it is damped, as we do mechanically with stuffing. Of course, the entire bass system, including woofer; enclosure, and crossover; must be designed to be used with this stuffing technique I now call Gradient Density stuffing. This results in what is called an aperiodic design, and can result in a loss of bass if not done properly with custom designed woofers for this use. To our knowledge, our exact system and its results have not been duplicated in other designs.

WSR Reber: Do you want to express any other views on subwoofers to our readers?

Von Schweikert: Yes, I believe in using multiple subwoofer enclosures in the room to smooth the response. Standing waves, caused by parallel walls, will be ameliorated by using multiple subs. This means the subs must be made unobtrusive, which dictates a compact design for most people. We have designed both our powered subwoofer, the Sub3 "Tower of Power" and the TS-3 full range rear surround, as "wall huggers." These are tall, wide, and thin, so that they can be placed against the wall. This design "disappears" in the room, and also allows the use of multiple small drivers (10-inch), which are far faster, cleaner; and more powerful than the typical paper 15-inch or 18-inch woofers used in old fashioned systems. The smaller drivers allow more silent space between notes to be heard, which larger woofers blur due to overhang. Each of the Sub 3's use twin 10-inch ultra heavy duty woofers with 2-inch high temp voice coils. Their cones are made of very thick polyplastic impregnated with carbon powder and are not only very rigid, but are light for transient speed. I designed the 800 watt amplifier around these drivers, with "intelligent circuit equalization" to extend the response to10Hz at cutoff.

WSR Reber: Do you recommend an additional dedicated subwoofer(s) to reproduce the ".1" low frequency effects LFE channel for maximum impact?

Von Schweikert: Well, that depends. I would normally say "yes" if I were a manufacturer of small compact mini-monitors that do not have adequate bass response on their own. Our main towers already have subwoofer response.

WSR Reber: I'm talking about the specific ".1" channel that's now part of the 5.1 channel format.

Von Schweikert: Of course, a powered sub or multiple subs are necessary for this channel for full effect. Hollywood has discovered that deep bass adds excitement, and I agree. As long as it's quality sound!

WSR Reber: I preferably use two sub-woofers on a ".1" channel, and then position them at different physical places in the room. But the ".1" is not always on, it's an on-and-off type of scenario depending on how film makers use it.

Von Schweikert: I agree that correct implementation of these bass effects can be really exciting. I'm afraid that many movie buffs have never heard good bass, since the theatres typically use outdated speaker systems.

WSR Reber: Do you think bi-wiring offers an audible improvement that justifies the cost? Can your speakers by easily bi-wired?

Von Schweikert: Our line of high-end audiophile systems use twin sets of binding posts for bi-wiring as we easily heard differences when using very high resolution sources and driving equipment. Home theatre sound, however; presents new challenges and opportunities. The benefits of multiple channels and newer; cleaner recording formats have opened up the sound to a degree never thought possible before. Bi-wiring five sets of speakers is a challenge, and does not seem as important as when only two speakers are played, probably due to the increased spatial cues and clarity associated with 5.1 channel play-back. For this reason, our Virtual Reality Theatre systems have been designed with peak clarity using only a single cable.

WSR Reber: How do you differentiate between your high-end speakers and your home theatre speakers? Because to me, using five of your high-end speakers would be optimal.

Von Schweikert: Well, I agree, but the price would be staggering for most buyers. Not only the cost of the speakers, but the double runs of cable to each speaker would be expensive, since you would naturally want to use the really good wire, too. This might cost as much as the speakers in some cases! This is like asking everyone to buy a Ferrari if they want a good car, because the engineering and cosmetics are extremely nice.

For practical reasons, in order to get people to put five speakers into their homes, we've downsized the speakers and eliminated bi-wiring. Most people won't spend the price of a nice car on their speakers and cables, so we have engineered our home theater systems to work at their optimum with a single cable. We have made advancements in engineering that allows us to do this, and they're not compromises. For instance, vocal intelligibility is enhanced by our proprietary woven carbon fiber drivers. Previously, we used to insist on bi-wiring to get that type of clarity in the midrange. It's simply no longer needed in this application.

WSR Reber: Your perspective in terms of absolute all out performance in a home theater system is based on cost considerations for the consumer; but do you agree that using four or five of your high-end speakers would be the ultimate experience?

Von Schweikert: Of course, I agree. Although this might be spoken from the Ivory Tower. I have set up five of our VR-4's in a no holds barred system, with an Angstrom processor and Balanced Audio Technology amplifiers, and the experience was devastating. No commercial movie theatre can even touch that sound!

Our new center channel speaker; the LCR-30, is a smaller version of the VR-4 designed for horizontal use. It is an all-out three way system with twin carbon filled 7-inch woofers, a 5.5-inch woven carbon fiber midrange, and a 1-inch aluminum dome "eyeball" tweeter surface mounted on the top of the array. Most people are not going to use a perforated screen, and hence need to use a horizontally placed center speaker. This is probably the best one available and is slightly expensive at $1,200. However; it has bass response down to 22Hz, has the highest clarity index I have measured, and plays as loud as almost any main speaker. A unique feature to our LCR-20 (a smaller two way) and the LCR-30, is the Vocal Presence control. This is a side mounted potentiometer that varies a new circuit I have designed to increase the vocal intelligibility when necessary. Some sound tracks have either a slightly muffled vocal track and need a presence boost, while others have the opposite problem. Harsh vocals and treble sounds can be tilted downwards with this control, enabling any soundtrack to be optimized. This is not just a tone control, it's a tilt control which can also serve as a timbre matching device when using another brand of left and right speakers.

WSR Reber: Do you believe that time and phase are audible parameters in loudspeakers? Do you favor slow or steep crossover slopes. Do you physically position the drive elements for time synchronization in your designs? Is diffraction important? What if anything do your designs use to affect diffraction?

Von Schweikert: This is a great set of questions, Gary. All of these topics relate to imaging and a three dimensional sonic recreation. First of all, I believe that time synchronized drivers and minimum phase errors in the crossover are just as important to imaging as proper elimination of diffraction. To begin with, we align the drivers for correct impulse response using purely physical offsets on ourVR3, VR4, and VR4.5 systems since they are large units. In the thin profile models like the VR1 000, VR2000, and VR3000, we are using passive analog delay circuits designed into the crossovers for the tweeters. Although this increases the parts count, complexity, and cost, the results are worth it, since we can't physically offset the tweeters in those designs without making them larger. Of course, the reason we offset the drivers physically or electrically is to ensure that the sound from the woofers, midrange, and tweeters arrive at your ears simultaneously If this is not done, the imaging will be two dimensional and lack "body."

Next, it is commonly known that first order crossover slopes offer both frequency and phase linearity that is not normally achievable with higher order designs. This is due to the phase shift caused by slight delays in the shunt circuits. However, there is a giant tradeoff to be made, since the very slow slopes will allow midrange and treble frequencies to be reproduced by the woofers, and bass/mid frequencies to be fed into delicate tweeters. The resulting four octave overlap causes several types of distortion, including frequency modulation of voices by the pumping motion of the woofer, mechanical distortion in the tweeter caused by excessive bass feed, and beaming/lobing effects caused by the vertical spacing of the drivers. Vocals being reproduced by woofers causes a "chesty" type of resonant coloration that I find quite disturbing, since it reduces intelligibility greatly However, the greatest problem is the inter driver beaming effects commonly referred to as "lobing."

If you have ever stood up while listening to a system using first order crossovers, you might have noticed a very large change in both the timbre and imaging. Twelve to eighteen dB dips and peaks can be measured in these systems at 45 degrees vertically off axis, since the path length to the drivers has changed drastically. With so much driver overlap, the target focal point has to be designed to focus at the spot the microphone is placed, usually on the tweeter axis. The distances from the mic to the drivers are all the same, and at this small spot, the system can be made quite coherent. However, moving only a few inches off axis causes the path lengths to differ, which unfocuses the array. It's just like unfocusing a camera lens or binoculars, exactly the same effect. These so called "phase coherent" systems using textbook first order crossovers are only accurate over a very small listening window, and measure and sound quite "colored" off axis. In addition, the power handling is reduced by 100 percent due to tweeter distortion. All of these problems bode ill for home theatre application as well as music.

To counter these problems, I have designed a new type of crossover circuit I call the Global Axis Integration Network. GAIN allows the drivers to radiate individually in spherical radiation patterns without driver overlap, and the steep slopes do not allow distortion or coloration to develop. Best of all, the topology allows phase consistency in a180 degree arc horizontally, with a rear firing ambiance driver filling in the rear 180 degree polar axis. The vertical phase consistency is also very good, allowing a listener to hear very little change as he or she stands or sits. Although the slopes approximate a 24dB per octave cutoff, this has been accomplished with harmonic filters which have no more parts in series than some highly equalized first order designs. Most of the work is done by Zobel parallel circuits which can be made phase coherent by clever mathematics and computerized circuit modeling. I don't recommend this to amateurs, as this has taken me a decade to perfect! In addition, the actual crossover points have required custom drivers, since I am basically using a full range 5.5inch woven carbon fiber driver with a response of 57Hz to11kHz, with augmented sub bass and super treble drivers added. This has the effect of sounding like a one way speaker, an electrostat, for example, since the crossover points are not audible and there is no inter driver distortion.

I have engineered the radiation pattern of the system to behave as a pulsating sphere or point source, which is the ideal, since it duplicates both the mic pickup pattern and the radiation of instruments themselves. This was achieved by measuring the off axis response of the total system, and then filling in the rear high frequency roll off with an additional rear mounted "ambience" driver. The signal fed to this driver is processed by a special circuit which simulates the pickup pattern at the rear of the mic. It is a phase and frequency compensated signal which totally complements the total output in a polar axis measurement.

Finally, diffraction is audible and must be eliminated from the speaker's output if clarity is important. It is simply reflected sound off of baffles which are too large, or improperly designed grill frames. These reflections interfere with the initial, direct wave from the drivers, and behave as sonic ghosts, just like visible ones on TV screens caused by either mountains or tall buildings. We use thick felt pads and minimum baffles on the VR3, VR4 andVR4.5 to eliminate diffraction, along with rounded edges. On the VR Theatre series, we use minimum baffles and rounded edges. There are no grill frames on any of our models, since they use cloth socks for cosmetic covering.

Home theatre was a driving force for these designs, since the days of a bachelor going down into his basement to listen to music alone are gone. This hobbyist would put up with weird speaker designs so directional that you had to sit very still, because if you moved a couple of inches, the sound would change drastically due to unintentional directional beaming caused by lobing effects and also by planar speakers which had no high frequency dispersion off axis. Of course, today the whole family is involved with home theatre, and the sweet spot has to be wide enough to fill the entire room with good quality sound. That's where our technology shines.

WSR Reber: At what physical distance in front of your loudspeakers do the various wave fronts from the drivers come into focus?

Von Schweikert: Approximately 1.5 meters, or five feet.

WSR Reber: Is that through your home theatre line as well as the high end line?

Von Schweikert: Right, all of them. The crossover slopes are steep enough to allow that, and the midrange and tweeters are placed in almost direct contact, allowing the short focal point.

WSR Reber: In your comments, I notice that you are distinguishing between a home theatre system and music reproduction systems in terms of subtleties of reproduction, so what is your view on the suitability of home theatre systems for music reproduction, and vice versa?

Von Schweikert: Well, I think that the same system that works for music will work for film, and vice versa. In other words, I don't believe the customer should have to buy two separate systems. There is no real reason why the sound of a film soundtrack has to be reproduced any differently than music, including the rear channels in the case of 5.1sources. It is possible to record music in a hall with five channels and feed that signal back into a five channel home theatre system which should sound like you are sitting in that very hall. Ideally, speakers should be accurate to the signal, and work well with the room, to sound like the original source. I also believe that music and film systems will merge in the very near future, and am working to make that happen. I imagine a sound quality that will beat any two channel system, regardless of cost. I do not advocate losing the high end ideal of transparency and purity and will not compromise that standard.

WSR Reber: Do you advocate using identical speakers at each position in a multi-channel system? I'm talking about an optimum scenario.

Von Schweikert: Absolutely, if a 5.1 discrete signal is being played. Especially if the recording engineers and mix down directors are aiming to achieve an accurate, true to life sound picture. A discrete stereo signal in the rear speakers will invalidate any reasons to do otherwise. The dipolar null designs that sound weird by themselves were designed to radiate ambiance from a monaural signal, and is a totally valid concept with Dolby ProLogic encoding. However, the discrete formats such as Dolby Digital and DTS Digital Surround invalidate the reasons for using non identical speakers in the rear. Herein lies the crux of the matter. Unless you use only one source format, you are going to have to buy a speaker system that is switchable in it's radiation pattern. Otherwise, you are going to compromise one format or the other.

In addition, I believe that the correct speaker to use in both cases would be one designed to replicate the output of the microphone, due to the encode/decode nature of recordings. That is what our systems do.

WSR Reber: I want to turn now to setup placement considerations. Do you recommend identical time distance relationships of each speaker from the sweet spot listening position in a multi-channel system? And I qualify that as a single sweet spot because when it all comes down to the bottom line, there is only one absolute sweet spot where a coherent multi-channel signal can converge in time and phase and amplitude level.

Von Schweikert: I believe that technically, in an old fashioned textbook sense, that is correct, especially when using conventional speaker systems other than ours. However, we have widened the sweet spot considerably with our Global Axis Integration design. This results in a wider placement flexibility and better dimensional recreation than conventional speakers. Let's make an example. Say that Indiana Jones is standing in a jungle with rain coming down, and that rain sounds like it is coming from everywhere above you. There is some guy creeping up behind him to get back his big ruby Indy stole from the statue. You, the observer/listener, are hiding somewhere in the bushes close by. You move several feet and hear a slightly different perspective, with rain hitting different trees making as lightly different sound. No matter where you move, you will be positioned differently than this person sneaking up on Indiana Jones. Let's say the microphone placement enabled you to hear these perspective differences with their different frequency and phase response patterns. Ideally, the perfect recording/loudspeaker interface would allow you to hear all of the above sonic information no matter where you moved in your listening room. Your movement would be akin to moving around the action taking place there in the jungle, with sound all around you. I think the sweet spot should be changed into a "sweet space" as large as the theatre room, and have attempted to design systems that will achieve this goal. This being a theoretical goal, you want to make sure that the speakers are not directional either vertically or horizontally, since any movement on your part will take you out of one speaker's signal path and into another's. Clearly, this is poor technology, but is what is commonly available.

Think of VR sound as being radiant like from an overhead light bulb, instead of beamy, like a flashlight. Clearly, one overhead bulb of sufficient brightness will illuminate a stage better than five flashlights, especially if you want to move around the room to gain different perspectives on the action. You can imagine how this radiation pattern frees up the placement problems and narrow sweet spots of conventional designs. I think as technology advances, the entire room will become holographic.

WSR Reber: I understand that scenario would work heifer using a point source microphone rather than multi-channel types of recordings or hundreds if not thousands of sound effects recordings mixed together to manufacturer a soundtrack. However, it all comes down to the mixing engineer's perspective, at a certain point they have to balance out the system from their perspective and this is a single sweet spot position that is the point from which the filmmakers create the intended sound field experience. I wonder how you would set up the system to establish a SPL level for each of the five speaker positions, because there has to be a desired spot for that meter to register level and for you to then adjust the appropriate equal level heard at that position. Typically that is a single sweet spot from the perspective of the recording or dubbing engineer who mixes all of the sound elements into a coherent acoustic reality, which matches the camera perspective of each particular scene. They are positioned and locked into one spot and they balance everything from that sweet spot perspective. That is why I brought up the time and distance issue, which I believe to be critical to locking in the full imaging potential of the mix. What I do is use a clock approach. A circle so that each speaker positioned on the arc is equidistant to the sweet spot. I then play with the fine positioning of the speakers on that arc until every speaker is time distant correct and produces the best possible stereo phantom imaging between any two pairs of speakers.

Von Schweikert: I totally agree with that. I want to get a little more detailed, I would advise using a tape measure or string to check the distances from the front three speakers. I have found that it is necessary to move the center channel speaker back a little in order to have equal distances. Most people incorrectly place the front three in a straight line, especially if they have wall unit or bookshelves. Ideally, you can use a decoder which has time delay built into the digital circuits, which really makes the setup easy. Then it doesn't matter it you can't place the speakers in a perfect arc. Many decoders on the market have this feature.

WSR Reber: Should speakers in a multi-channel system have identical full range capabilities for each discrete channel?

Von Schweikert: Absolutely. They should all be flat down to at least 80Hz if you are going to be using a subwoofer augmented system crossed over at that frequency. However, being flat down to 80Hz is not the same as having lot's of air movement and power at100Hz. Many mini speakers have response down there, but absolutely no air moving capability or power handling, so it's not going to sound as real as a larger, more powerful system. The mid bass/lower midrange area is critical, and minis can't hit that range with enough power. I am greatly concerned about miniature center and surround speakers as well, since there seems to be a tendency for designers to make overly small systems based on cosmetic considerations. They look great, but sound weak, tending towards thinness and brightness. Pans from large mains to the center sound weird since the lower mids on the tiny center do not sound anything like the mains, which are fuller and richer. Same with pans from front to rear, sounding very "tinny" in the back channel compared to full in the front mains. This potential and common problem forced me to design very powerful 5.5inch carbon fiber drivers capable of handling far more power than average units in the bass and lower midrange. Our small center channel LCR10 and LCR20 sound like much larger systems, as does our small TS100 and_TS200 wall mounted surrounds. They have full power output at 80Hz without distortion and brightness and match large mains in both power handling and timbre matching. Those looking for compact size will be pleased.

However, if we're talking all out performance, matched large systems are some ½hat better if they all go down to the 20Hzrange, like our LCR30 and TS300 systems ½ or center and surround.

WSR Reber: Yes, to cover the octaves down to 20Hz and then a dedicated ".1" subwoofer in addition to that for the special effects.

Von Schweikert: Very few systems available in the chain stores can provide that capability, since they can't even get timbre matching right.

WSR Reber: Do you subscribe to the Home THX™ specification for dissimilar speakers using direct radiators in the front hemisphere of the room and dipoles in the rear hemisphere of a multi-channel system?

Von Schweikert: I agree with THX's rear channel dipole philosophy as long as we're using the older Dolby ProLogic format with monaural rear channel information. The dipole null reduces the amount of speakers you need around the sides and rear of the room to provide ambience from a monaural source. However, this is not going to work with stereo signals available with 5.1 digital formats, since we need to steer the rear signals instead of simply spreading them around the room. Although THX™ believes that dipoles will work with stereo in the rear, I and the majority of other "purist" designers and theoreticians don't agree. However, the bulk of programming will continue to be Dolby ProLogic with almost 15,000 titles, so dipoles will be around for a long time. Still, anyone setting up a home theatre system today should look at surrounds that can be used for both diffusion and localization. There aren't many on the market yet, but I believe that type of design will eventually predominate. I had a working prototype two years ago but wasn't able to put it into production until recently due to lack of available program material in the 5.1 discrete format.

As far as their obsession with vertical directivity control in the front LCR's, it is only really necessary in highly reverberant rooms such as commercial movie houses with concrete floors and high amounts of echo. Tomlinson Holman found that high amounts of reverb reduced the vocal intelligibility, and solved that problem by not allowing the LCR's to radiate treble in a wide vertical arc that would excite the floor/ceiling boundaries. Normal living rooms don't have those same problems and don't benefit as much from the vertical directivity design. In fact, many designers and reviewers don't like the way they sound on music, and I have found that this is due to nonconformance to natural radiation patterns found in real life. Some also are slightly too bright due to multiple tweeters. However, to their credit, THX™ was among the first to specidentical speakers for the front three LCR's for timbre matching reasons.

WSR Reber: Controlled vertical dispersion was a practice done from the very beginning in theatre sound. Not to Tomlinson's credit, but something that's been part of the design element of horn loaded systems in movie theatres. It's controlled dispersion.

Von Schweikert: Two recent changes I have seen in THX™ philosophy involve a horn loaded dome tweeter, and a three way center 3channel using a single tweeter centered between two small midrange drivers. Both of these systems have received good press and have evidently eliminated the gross prior problems. Being a designer of THX™ systems for other companies, I have worked with Lucasfilm extensively and find that they are evolving constantly. Anthony Grimani has a very open mind and is a pleasure to work with. Holmanhas left the company, by the way, to start his own speaker company making studio monitors.

WSR Reber: Are your designs THX™ certified?

Von Schweikert: Some of my earlier designs have been, like the Counterpoint Home Cinema system that won the Best Product of1995 (speakers over $2,500) by Video Magazine. I am currently working on some designs that will be THX™ certified for an over seas company However, these are not like the older THX™ designs but are departures with a very musical sound, and front to rear timbre matching. I am also working on a very new surround design that no other company has dreamt of.

WSR Reber: Are we talking about Von Schweikert Research products?

Von Schweikert: Yes, a new type of surround system with automatic switching for the rear radiation pattern. The system will eliminate the problems experienced with monaural signals.

WSR Reber: I have never had that experience with monaural surround.

Von Schweikert: Really?

WSR Reber: Yes. Not if you set up your system with equal time distance with identically matched speakers and you are at the typical sweet spot, and the surround speakers are not directly to the sides of your head. I agree that if the direct radiators are position directly to the sides of your head as are the dipoles with the null to your head that the dipole will work better, but when you position the identical to front surround speakers on the arc so that they create an optimum phantom side wall image, I have found the imaging quality is superior, especially with discrete 5.1 and Fosgate Six Axis or Circle Surround 5.2.5 matrix processed material, which is a superior experience to Dolby ProLogic in any event.

Von Schweikert: Well, in one study I conducted, THX™ personnel came to the factory and we set up two small speakers, one on each side of our head, and sat directly in between. The problem was obvious.

WSR Reber: I agree, it should be. But the problem is the setup. If you have them firing directly into each ear, yeah, then you are going to hear the speakers localize.

Von Schweikert: Well, that's what they are predicting that the customer might do.

WSR Reber: The real problem is lack of education to realize an optimum setup. If you have your speakers equal distance and you place your right front and right back (or left front and left back) so that if you put a stereo signal on that wall you would have a solid phantom image, you won't have that problem. That is my experience in the experiments we've done.

Von Schweikert: I think that one of the problems that THX™ was trying to circumvent was the fact that they weren't really sure where the customer was going to place the speakers. Although they recommend an elevated speaker placement above ear level, most people ignore that fact due to room décor or inability to read instruction manuals. So I have pioneered a new type of surround speaker that has several different types of dispersion patterns and will operate correctly with any type of input signal. Monaural signals need diffusion in many applications in order to achieve room filling ambiance, yet the newer discrete formats need the opposite type of pattern which is directional. Also, there may be optimum amounts of both types of dispersion patterns to simultaneously radiate both diffuse ambience and steerable signals at the same time, and apparent dichotomy

WSR Reber: Well, where does that presume that the two left and right surround speakers are positioned relative to the front speakers? Because if the speakers are placed directly to the sides of your head and are played in the direct dispersion mode you will have the same problem as you found with your experiments.

Von Schweikert: The surrounds can be placed to the sides or in rear corners, since the directivity is adjustable. The idea is to make a speaker which is flexible from the customer's viewpoint.

WSR Reber: My first thought, would it violates the principle of equal distance placement and the principle of optimizing for side wall phantom center imaging.

Von Schweikert: This is a general concept, and we're working on placement flexibility. This is tied into processor enhancements using time delays as well, to ease placement requirements. Nothing is perfect, even the best equipment sold today is an approximation, and I am trying to make the approximations sound a little bit more realistic.

WSR Reber: Well, on a processor; you can work out the challenge of equal time distancing your speakers to one set of ears, but you can't work out the design challenge of phantom imaging between a left back and right back speaker relative to their corresponding fronts. You could move the left back speaker up to be three feet from the left front and still not have good phantom imaging. There is certain distance, just like you have a certain distance between the left front and the right front stereo perspective to get a good phantom image.

Von Schweikert: So what we're talking about concerns optimum placement.

WSR Reber: Yes. Your design allows the speaker to change dispersion characteristics?

Von Schweikert: Right. It's basically a tripole speaker that has an ambiance system radiating from the sides and a directional system radiating from the front.

WSR Reber: And can they play simultaneously?

Von Schweikert: There are switches on the speaker that enable a selection between a straight dipolar THX™ null or a normally directed front radiating system matching the front speaker's dispersion pattern. The third switch position allows a combination of both directional information and a derived ambiance signal radiated from the sides at a lower volume level, in order to achieve a new type of envelopment. This is adjustable and is more or less an effect that may or may not be used, depending on the room, the placement, the source, and the individual's listening preferences. You use this if you really want to "open" up your room. It certainly sounds better than mono! Many rooms I've set up did not have adequate envelopment of the rear channel information due to lack of symmetry or décor problem that would not allow placement in a clock arc for all five speakers. The front stage would sound great, but the rear fill was not as powerful or all encompassing due to the use of small rear speakers and lack of a suitable place to put rear speakers symmetrically. Most people don't have a dedicated, symmetrical room for their theatre system, and my design attempts to correct problems in the real world. Since these effects I described can be switched out, the system can be used as a conventional speaker for those rooms that allow it using correctly recorded source material.

WSR Reber: Your surround speakers then are dissimilar to the front speakers, they are not identical?

Von Schweikert: They can be made identical with the switching facility I mentioned. The effects can be turned off at will.

WSR Reber: So you have a directional speaker system but that means to optimize for the THX™ dipole null dispersion it is firing into your left and right ears?

Von Schweikert: Not necessarily. It depends on where you place them. Flexibility is the keyword here.

WSR Reber: When they place them on aside wall, I don't see people moving them to the rear when they want to change experiences. They're probably going to place them in one position and leave them there, so they are probably going to put them in the side wall because that's where THX™ recommends they be put, so you are going to have basically a direct firing speaker into your left ear and right ear; complimented by, at your choice, the other two speakers to be used as dipoles out of phase, or in phase. So now you've got three speakers in phase together.

Von Schweikert: Well, part of the effect consists of an ambiance signal sent into the left and right systems in order to avoid what would turn into an interference pattern. You know, if you have three speakers that are all aiming in different directions, you get a comb filter and ripple effect caused by driver overlap. To eliminate the problem you are talking about, I am feeding a "difference" signal into the drivers on the right and left side of the unit. This circuit also compensates for the problem caused by side wall placement as opposed to the rear room placement preference. The customer is able to dial in ambiance that won't interfere with the directional signal if necessary This eliminates the problem of the speaker playing directly into the right and left ears if placement is on the sides of the room.

WSR Reber: So that ambiance signal is derived from the signal going into the speaker by a processor box?

Von Schweikert: Yes.

WSR Reber: Then basically you go through some processor box which is able to distinguish how you want to choose the signal application or processing so you can have out of phase material from that signal going to the two side speakers and then the one facing your ears can be in phase.

Von Schweikert: Right, exactly. This is a special effects system which will give you a bigger; fatter sound back there, more envelopment if you need it. If a purist doesn't need it, it can be shut off.

WSR Reber: And use just the single speaker.

Von Schweikert: Right. The reason we developed that system is because when you hear a dipole playing a discrete soundtrack...

WSR Reber: The dipoles loose it.

Von Schweikert: Yes, and all of a sudden they are thinking they don't want that null anymore, but what are they going to do? We thought, why not build a speaker that will not be obsolete due to source material and format changes.

There are a lot of ways to envelope the rear of the room and a lot of people have strong opinions on how to do it correctly. Bipolar, dipolar; and conventional front firing systems have their adherents and coexist in the market place. We are trying to make a system that is very flexible and will please almost everyone since they can adjust it to provide what they want from the surround experience. I believe this flexibility to be a better option than providing just one radiation pattern which the customer is stuck with.

WSR Reber: I know of music projects recorded for discrete multi-channel and have personally produced a fair amount of projects over the years. I've always used identically matched speakers and used the clock approach to achieve a solid phantom image on all four walls. That's how music is being mixed now for DTS Digital Surround encoded CD's.

Von Schweikert: I agree, and that's why these effects can be switched on and off.

WSR Reber: You make very high-end loudspeakers. What's the VR4?

Von Schweikert: The VR4 stands for virtual reality in four dimensional planes: frequency phase, time, and space. This $3,650 system has been favorably compared to other systems by several reviewers.

WSR Reber: Albert, what would you personally choose, five identical VR4's or your home theatre system you've just been describing, with the narrow profile VR3000 up front, the LCR30 center; and the specialized IS300 effects surround speaker with adjustable radiation patterns?

Von Schweikert: If I started with a discrete 5.1 digital system and didn't have any old matrix Dolby ProLogic laserdiscs, and had unlimited financial resources, I would naturally go with the more expensive VR4's, even going to the extent of putting one behind a perforated screen. My room is 35 x 45 feet and requires a lot of sound output. The home theatre package is designed for smaller rooms and for the person with multiple source formats. It also costs about half the price of fiveVR4's, but I'll bet that the average person wouldn't notice the differences since he or she would be watching the movie instead of criticizing the sound.

WSR Reber: Is a different model permissible for the center channel?

Von Schweikert: Theoretically, no. You definitely want the same type of sound coming from the center; and it also happens to be the most important speaker in the entire system. It not only has to handle voices with clarity, but music and special effects at the same time, with explosions, storms, loud music, and what not, all trying to compete with the vocal intelligibility. In some soundtracks, way over half of the volume level is directed at the center speaker, with only occasional pans to the main left and right. Many manufacturers of large main speakers have coloration problems in their lower midrange frequencies caused by vocals being reproduced by large woofers with inadequate speed and articulation. To compensate for the "chesty" and resonant sounding vocals from the mains, they "voice" their center channel speaker with a leaner; thinner sound, a bi-product of smaller drivers. This does enable better vocal clarity by eliminating the chesty resonance, but now pans from the right or left side going across the center aren't matched. For example, trucks driving by don't sound right when they go past the center, as the deep engine tone is gone momentarily. Other worse problems occur with people walking across the room while speaking. The large timbre shifts in human voices are very distracting to the all but the most casual of listeners.

Years ago, at the lab at Cal Tech, I did research on what I now term the "clarity index" of different types of drivers. Cone flex occurring in paper or soft plastic cones, such as polypropylene, absorbs minute transient details in the signal that contribute to the leading edge of spoken words.

We used a twin tone test, wherein we superimposed a quiet tone on top of a louder one, varying in frequency. We measured almost complete cancellations at some combinations of frequencies due to the cone flex, which we later were able to see visually. Using a strobe light and a calibrated signal generator in a dark room, we could see the cone appear to be stopped, much as a timing light works its magic on the flywheel mark when timing an automobile engine. Polypropylene and doped paper cones are very prone to flex induced colorations and are only used by cost conscious designers attempting to hit target price points.

Clarity, or the lack of, is not even a factor in some cases. I consequently have been developing exotic cone materials for the last decade. I was a driver designer at the world's second largest manufacturer of raw speaker parts for two years and worked with Kevlar, woven carbon fiber, and ceramic materials. The best Young's Modulus, or ratio of lightweight to stiffness, is exhibited by woven carbon fiber threads. The resulting fabric is stiffened with resins, and tiny traps result from the weave. Standing waves cannot form on the cone due to the lack of a contiguous surface, and the cone is too stiff to flex. The light weight allows for much faster transient response as well. The combination of the factors add up to a high clarity index, approximately seven times better than poly or paper cones. Thus, using these special cones, we are able to achieve a very high level of vocal intelligibility without having to resort to voicing tricks such as leaning out the frequency response in the lower midrange. All but our base priced systems use this midrange cone, and use aluminum tweeters as well. Fabric tweeters have diaphragm flex worse than the midrange cones since they operate at higher speeds and are much more flexible. It is amazing how much vocal harmonic information is reproduced by the tweeter at 2kHz. Aluminum domes have a far higher clarity index than fabric, and can handle more power due to heat transfer from the voice coil to the dome surface. The combination of our exotic drivers and Global Axis Integration Networks result in a state of vocal intelligibility from all our systems. Along with the fact that we're not feeding vocal information into the woofers, all of our speakers are extremely clear and match in timbre.

WSR Reber: Is this center channel speaker identical to the left and right front or is it positioned horizontally?

Von Schweikert: It's horizontally placed, but is identical otherwise. The largest model, the LCR30, has twin 7inch woofers, a 5.5inch carbon fiber midrange, and a surface mounted aluminum "eyeball" tweeter Although somewhat large at 27inches wide, it is perfect for 32inch direct view sets or large rear projection units. Our smaller system, the LCR20, uses twin 5.5inch carbon fiber woofer mids with a 1inch aluminum tweeter and is only 19inches wide. Both of these are aperiodically vented systems with very deep but very clean bass response. A unique feature of both models is the Vocal Presence control, which not only works for timbre matching purposes, but will enable the user to adjust the tonal quality of the midrange frequencies. Voices can be either pulled up or pushed down into the mix to control sibilant or muffled recordings.

WSR Reber: What about the impact of the larger baffle surface created by the screen itself?

Von Schweikert: We used a large screen unit when designing the radiation pattern and frequency response. We didn't design it in ananechoic chamber, since that would result in false operating conditions. The reflections of the screen have been factored into the design, but if the location is changed, to perhaps floor or ceiling mount, the response can be changed by the Vocal Presence control. This control has settings which also can tilt the frequency response to compensate for differing bass boost conditions caused by the environment.

WSR Reber: How do you address diffraction in both left and right fronts as well as the center channels?

Von Schweikert: We are using minimum baffle designs with very little surface area to cause diffraction. In addition, all cabinet surfaces are radiused 1inch, and there are no grill frames, all units being covered by a sock. The largest center, the LCR30, has no baffle whatsoever for the tweeter as it is mounted in free air.

WSR Reber: In addition to test instruments, what source material do you use to evaluate loudspeakers? Music? What do you recommend to consumers?

Von Schweikert: The ultimate test may be difficult for consumers to duplicate, but this involves feeding a live signal from a voice or acoustic instrument through the speaker. You need an instrument grade mic, a mic preamp, excellent cables, and a good amplifier. The idea is to listen to the instrument, then pipe the sound directly into the speaker and compare the two. When we did this in 1976, it was a humbling experience. This direct feed testing allowed us to quickly assess problems and make correct value judgements. Using recordings is not good since you have no way of knowing what the original sounded like unless you were there. Even then, your acoustic memory is not so hot. As I recall, our first speaker sounded nothing like a person's voice, although it measured good. We were using a B&K gliding tone generator, chart recorder, and making the measurements on the tweeter axis, just like everyone else. We soon came to conclude that off axis response measurements, such as the polar axis we now use, are far more indicative of what the speaker is going to sound like. We can come surprisingly close to a live source.

WSR Reber: What does the consumer do at a retailer to determine speaker sound quality?

Von Schweikert: Consumers should get a pink noise test disc and learn to listen for dips and peaks in the response. Get the sales man to compare different speakers you are interested in. No colorations can hide from pink noise! If there is a nasal quality to the midrange, you'll hear it instantly, while with music it may take you much longer If the tweeter is screechy you'll hear that instantly too. Pink noise sounds smooth and has only one tone. If you hear something sticking out, it is a coloration that you don't want. Also, try sitting and standing while listening to the tone. You can test for lobing and directional problems, since vertical movements will expose crossover flaws and driver radiation problems instantly Walk around the room, listening for gross changes in the frequency balance and centered image of the pink noise "ball," for that's what it sounds like if the speakers are hooked up properly in phase. If something is hooked up improperly or if one of the drivers has been accidentally wired wrong at the factory the image "ball" will appear to be diffuse and will not float in between the speakers. To train yourself, listen to pink noise at home, and reverse the leads on one of your speakers. You will be able to learn what to listen for and will become a better trained listener as a result.

Us ten to acoustic recordings of simple instruments such as acoustic guitar and voice, and listen to voices on soundtracks that are well recorded. Sound effects like car horns, trucks driving by and other neat things can train you to listen for natural sound quality as well.

Read good magazines, like Widescreen Review to learn more about technologies and products and how they stack up against others. In other words, listen to somebody who knows what good sound is because they've done testing on a professional level. Read the product reviews to get basic information.

WSR Reber: Never trust only the product review, always go and experience it yourself.

Von Schweikert: Consumers may not immediately like a really accurate speaker. They might like something initially that's bass heavy. Of course, they should buy something they genuinely like, although I suggest they buy something accurate and learn to appreciate it rather than buying a bass heavy model and hating it in a few months.

WSR Reber: Any other comments?

Von Schweikert: I personally believe that the future of audio in general, not just home theatre, is going to go to multiple channels as the 5.1 system. Two channel is going to relegated for Walkmen and headphone use on trips and whatever. Since playback will going to multi-channel, loudspeakers should be designed and optimized for this use. so, music and soundtracks should be equally well reproduced, since theoretically there shouldn't be any differences.

Gary, I applaud what you are doing letting the designers to focus on the important things. I'm designing for complete realism, whether for music or film, and I would hope that multi-channel reproduction becomes standard. I think we're both headed in the same direction.

WSR Reber: Thank you Albert for an insightful interview.

by Gary Reber
Widescreen Review
Volume 6 Number 1