|Due to working on new loudspeaker models, Albert Von Schweikert informed us that he sadly wasn't able to respond to our three simple questions of last month's first part of this two-part expose. Alas, perusing his website 30 days later, the astute reader wouldn't have failed to notice certain changes. The peculiar inverse cardioid dispersion pattern claimed previously as resultant from Albert's so-called G.A.I.N. circuit has vanished. It's now described as " our Global Axis Integration Network crossover design and driver radiation pattern allows the speaker to radiate a "bubble" of sound instead of a narrow beam, allowing 3-D imaging over a very wide space" [underlined phrases for emphasis].
Whatever happened to the heart shape featuring prominently in Albert's propaganda until recently? This verbal adjustment now duplicates the well-known broad dispersion ideal of the Canadian NRC's research. It's thus not novel but does conform with an established school of thought. Albert merely replaced the popular terms broad dispersion or wide off-axis response with "bubble of sound". He thereby also responded slyly to our earlier assertion that a network per se cannot arbitrarily steer sound into specific dispersion shapes. Rather, dispersion is primarily a function of driver radiation. It occurs post-crossover as a function of diaphragm size, shape and cone/dome/panel geometry. Dispersion is secondarily contingent on the exact bandwidth which a specific driver is asked to reproduce. This is clearly controlled by the network's cut-off frequencies and rate of attenuation. Physical driver spacing and the amount of bandwidth overlap -- i.e. how many notes are jointly reproduced by two or more drivers -- create lobing or interference patterns. These are standard challenges for good loudspeaker designers who attempt to minimize them in their crossovers.
|We now read about a "phase-consistent Global Axis Integration Network with stacked First Order circuits configured to enable 4th order acoustic slopes for minimum crosstalk distortion and reduced lobing" [underlined phrases for emphasis].
This appears to be a revised version of the following paragraph we quoted in Part I: "In effect, the Global Axis Integration Network employs ladders of first order filters, ensuring purity and coherence. However, the use of several stages of first-order filters also enables our GAIN circuit to have the benefits of fourth order filters, with lowered distortion and greatly improved off-axis response behavior [underlined phrases for emphasis]."
The terms phase- and time coherence, when employed according to their technically correct conventions, refer exclusively to the minimum phase shift and concomitant minimum time delay of 1st-order series or parallel networks. It's something we pointed out at length in Part I. We now observe Von Schweikert's claim for the G.A.I.N. circuit amended - from phase coherence to phase consistency. Since this terminology, though impressive-sounding, fails to be part and parcel of standard textbook speaker physics, we don't know what type of electrical behavior it attempts to describe. What is clear as daylight? That steeper 4th-order attenuation rates create less inter-driver overlap and thus less potential lobing problems. Hence the new claim for reduced lobing is factual. Give it up for common-sense speaker building 101. Alas, it's about as ordinary in this context as stating, with implied gravitas, that a tweeter reproduces high frequencies.
The term crosstalk distortion usually describes signal leakage or interference between channels. It's not a customary synonym for loudspeaker lobing. Granted, a case could be made that lobing (interference from adjacent drivers) equaled crosstalk (interference from adjacent channels). However, an opposite case could also be made. This type of ambidextrous verbosity attempts to infer the removal of a specific kind of distortion most audiophiles recognize as extremely undesirable. However, this particular kind of distortion plainly doesn't exist in speakers. After all, to suffer crosstalk requires a minimum of two channels. A single speaker's lobing challenges arise in a monaural context. If compelled thusly, one could imply that Von Schweikert was claiming victories where none were to be found.
Fancy terms undergo further liberties. We read that "the Von Schweikert Global Axis Integration Network circuit enables complete phase integration between these outstanding drivers, allowing a natural and consistent blend of midrange frequencies from the woofer and tweeter [underlined phrases for emphasis]." Like phase consistency, phase integration sounds impressive but has no technical precedents or proper accompanying definitions to explain what's meant. These terms are thus essentially meaningless yet liable to impress those not familiar with the correct terms as they are commonly used.
Seemingly intent on saving his famous microphone tie-in from lunar ruin, Von Schweikert then reminds us that we should "realize that the goal of the speaker should be to behave as a microphone in reverse. Of course, there are several types of microphones, but they all have common but important characteristics which the speaker must emulate:
Point-source behavior: mics receive the sound waves upon a very small diaphragm (12mm to 26mm in diameter). In order to project the sound back into the listening room in the same manner as the mic received the signal, a correctly designed speaker must operate as a point-source; this entails using a focused array of drivers in order to reproduce the full frequency bandwidth" [underlined phrases for emphasis].
Proponents of line source theory (Nearfield Acoustic, SoundLab, Magnepan, MartinLogan) would naturally disagree. Adherents of concentric symmetrical arrays (Dunlavy, PBN, Dynaudio) might have difficulties recognizinig a focused array of drivers in the VR-2 described here. Regardless, one notes that the previous claim for the inverse cardioid pick-up pattern is now abandoned with the admission that there are several types of microphones (with distinctly different pickup patterns). The focus has shifted to point source behavior, with the physical distance of the VR2's front- and rear-firing drivers making for a rather large 'point'. Also, does that imply that if more than two microphones were used during the recording, one now requires the equivalent number of point-source speakers to properly reproduce such recordings?
|"Coherent driver integration: As the recording mic is composed of a single diaphragm, it is by definition phase-coherent. In order for a multiple-transducer speaker system to behave as a focused point-source array, the drivers all must be in phase, so that the acoustic pressure mimics a one-way system using a single diaphragm. Forcing multiple drivers to start and stop together is a formidable task, compounded by the fact that the woofers and tweeters have different masses, suspension compliances, and sensitivities. Our proprietary Global Axis Integration Network corrects the driver responses, ensuring coherent sound and a thrilling multi-dimensional sound stage" [underlined phrases for emphasis].
|Here we have a beautiful example of obfuscation. Most audiophiles are familiar with the terms time- and phase coherence since 1st-order proponents like Thiel and Vandersteen wield them as arguments for their design philosophies. However, these terms are not synonymous with "being in phase". In this context, in-phase simply means that the drivers are not connected 180 degrees out-of-phase. Alas, two in-phase drivers receiving their signal after 360 degrees of 4th-order phase rotation are not phase-coherent but delayed in time by a full cycle. Dependant on frequency, this creates not one fixed value of time delay but an infinite number (see Part I for the accompanying math). Note that Von Schweikert doesn't outright claim time or phase coherence. He applies more finesse to his misdirection. He correctly calls a single-diaphragm microphone phase-coherent.
He then only implies, in context, that a point-source array, by mimicking the microphone, will duplicate its phase coherence as well. Clever. He then reinforces this implication by calling the results coherent sound. While not violating the laws of Physics as his quotes in Part I were so clearly guilty of, these descriptions should still be accused of deliberately implying 1st-order benefits for higher-order networks. You're either an orange or an apple. Von Schweikert would have you believe that he has miraculously cloned an appange or orapple. Hmm. Crabapple is probably more like it - small'n'sour in real life and non-existent if spelled Crap Apple.
Further in the 1st-order phase & time domain, Von Schweikert makes the following claims: "Since a wide-bandwidth speaker system must employ more than one transducer to cover the entire frequency range, the blending of the drive units is critical to develop coherent sound waves that will result in a three dimensional sound image. Many speaker systems do not use phase-consistent technology due to the difficulty of starting and stopping all of the drivers in unison. Due to the differing cone/dome weights, suspension stiffness, and efficiency, most multi-band speaker systems cannot be made pulse-consistent, but worse, many designers utilize crossover designs that necessitate out-of-phase driver wiring! Note that a poor sound-stage image is developed with these incorrect techniques, along with poor timbral response. The VR series of designs, including the VR-2, utilize specialized transducers that were engineered to operate together in unison. To eliminate complex circuitry that degrades detail, the transducers were designed to eliminate peaks in their breakup region. To ensure that the four transducers in the VR-2 speaker start and stop in unison, the diaphragms and motors were developed as a complete package, and in combination with our proprietary Global Axis Integration Network, the consistency of the sound pressure is very uniform and faithfully accurate to the signal. In fact, there is less than 3 ms of total delay between the transient impulse of the woofer, bass/mid and tweeter of the VR-2 system, ensuring a highly dimensional sound stage and correct timbral balance between the transducers. This low level of transient delay is below the critical auditory threshold discovered by the leading psychoacoustic researchers, and is a major factor in the realism generated by the VR-2 speaker system" [underlined phrases for emphasis].
Besides belaboring the obvious -- that every driver manufacturer in the book designs his transducers with minimal breakup modes in the passband the driver is intended to be used -- Von Schweikert again engages in liberally intermixing terms from the phase- and time-coherent brigade of loudspeaker design. He creates new ones that seem to be saying one thing while never being properly defined to become specific: "Coherent sound waves": "phase-consistent technology"; "consistency of sound pressure". In fact, the latter's implied meaning does have a proper textbook equivalent - power response. But it isn't used - as though acknowledging proper engineering terms was too common for AVS speakers. Instead, phase consistency becomes a completely made-up term. It sounds good but doesn't mean a thing - unless Von Schweikert were to define it succinctly. The absence of such definitions is as telling as the use of vacuous terminology to begin with.
The strangely insistent tie-in with microphones continues. "Microphones have far less distortion and coloration than average speaker systems, so the reduction of distortion and coloration is paramount." The fact is, every single component preceding the loudspeaker in the playback chain suffers less distortion than the speaker which transforms electrical impulses into mechanical motion. Attempts at reducing loudspeaker distortion are mandatory for High Fidelity, true - but not for the exact reason given.
"The best mics have a frequency response of 20-20kHz. This wide bandwidth is essential if the scale of the music is to be reproduced at full dynamics, especially in the lowest octave." Now we're led to believe that the reason a speaker must reproduce wide bandwidth has to do with microphones when microphones and all other audio-related components go after this particular bandwidth simply because it equates with how far human hearing extends in both directions.
It's as though to appear unique and armed with breakthrough technologies, Von Schweikert remained stubbornly married to his marketing spin that the perfect loudspeaker should act as a microphone in reverse, and that his are uniquely empowered to do so. If you find yourself unwittingly in agreement with the original premise -- it sounds as good as the old myth of the component footer as uni-directional drain valve for vibration -- you're nearly bound to agree with his conclusion as well. Albert clearly has not contested our earlier proof that his prior core claim for the inverse replication of the cardioid pickup pattern was bogus. He simply amended the respective terminology to the far more vague 'bubble of sound' that no longer corresponds to a specific shape, thus admitting that it was a false claim to begin with. Makes you wonder what else he's cavalier about?
As shown above, he's also minimally adjusted other areas of his claims which we've shown to be fake last month. So he's no longer in outright defiance of common sense and textbook physics. Alas and in my book, he's still guilty of shearing the sheep that are his prospective customers. The fact that he has made certain adjustments in the wake of our article is clearly suggestive. The fact that in 30 days, he hasn't been able to respond to our very simple three earlier questions?
After all, we didn't ask for lengthy treatises but short to-the-point explanations, something any expert in this field should be expected to rattle off coherently and in five minutes while making dinner.
|In the final analysis, every maker of audio components attempts to present his products in a unique light, to differentiate them from the competition and make you curious about their performance. If Albert Von Schweikert perhaps goes further than most, many would say that he's not only well within his rights but should be commended for audacity and creativity. Perhaps so. My personal feelings run more along the lines of believing that this undermines not only personal credibility but also the credibility of High-End audio at large. After all, why promote BS if you're for real? Why adjust terminology in the wake of being questioned when it wasn't BS to begin with? This simply smacks of a con job which, while not breaking any man-made laws but purely basic Physics, conflicts with common sense, grace and good taste. The alarming thing? How many industry writers seem to embrace such rubbish as engineering genius. Makes you wonder whether somebody opening up the G.A.I.N. circuit wouldn't, in the wake of publishing its schematic, observe similar adjustments of claims as we've described today. Would G.A.I.N. disintegrate as magically as the cardioid crap? Be that as it may - the prosecution now rests its case to let others design loudspeakers and describe their design goals and operation as they see fit.
Me? I'm looking forward to shortly reviewing Roy Johnson's Green Mountain Audio Continuum loudspeakers. They fall squarely into the phase- and time minimalist network camp yet refuse to make questionable claims of incorporating 4th-order design advantages. If you subscribe to one specific design school, why bother being all things to all people? Call a broom a broom. Let HP aka Harry Potter decide whether it flies. Or is it simply bad muggle medicine, useful only for cleaning dirty floors and cloudy thinking? For more hilariously bad medicine dispensed by the same audio pharmacist, click here.