SQ Shadow Vector Soundfield Mapping

[Lynn Olson]

We have to do a deep dive into what's called "pairwise mixing", which is what was common back in the quad days. Basically, to do a circular pan, you mix a one pair of channels at a time ... if the pan is clockwise, you pan from LF to RF, then RF to RB, then RB to LB, then LB to RF, and keep going if desired. If a joystick is used, then more than 2 channels may be present for some portions of the pan.

The gotcha with "discrete" recordings is that pairwise mixing doesn't necessarily deliver the most natural impression in the rearward 270-degree arc. It's fine for the frontal arc, from LF to RF, but kind of falls apart on the sides or rear. You can test this for yourself by routing a normal stereo recording between LF and LB, or RF and RB. If you turn 90 degrees and face the speaker pair, the image will only be mediocre if the speakers are dissimilar (crossover slopes and frequencies don't match), or if the heights are different. Then face forward and notice the image quality gets even worse (because the ear doesn't have very good localization for sounds coming from the side, which is why it's instinctive to face towards a new sound). Typically, the phantom-images between the pair will be diffuse and uncertain, and lower quality than the frontal phantom images. It's very common in consumer-grade surround systems for no phantoms to form at all, leaving a localizations pinned to either front or rear speakers, with nothing except a vague wash of sound in-between.

The same applies with greater force to the rear arc. In fact, a real, physical sound from directly behind you can sound very much like a sound directly in front, and if this is a phantom image created by LB and RB speakers, most likely it will have no apparent localization at all, unless you turn around 180 degrees and face the speaker pair.

The implication is that even "discrete" recordings will have vague localization in the rear 270 quadrant unless special mixing techniques are used, for example, Ambisonic's deliberate introduction of carefully phased crosstalk. This is another way of saying that pairwise mixing is satisfactory for conventional stereo (although not optimum) but only fair for quadraphonic or 5-channel.

So a conventional pairwise-mixed discrete recording should not be the gold standard for a matrix system, because it has a somewhat unnatural sound as the localization is swept through the full circle. The reference standard for any system, discrete or matrix, should be reality itself, not another recording system.

If ... and this is a big if ... a matrix system is carefully designed and artifact-free (this is pretty rare), it can sound better than a conventional pairwise-mixed discrete recording. For my own system back in the day, I always preferred the Shadow Vector decoding to a not-so-great CD-4 disc recording, but that was partly because of the low fidelity of the JVC CD-4 system, and show-offy "discrete" mixes that tended to localize sounds right at the corners ... the quad equivalent of "ping-pong" recordings. For some reason, CD-4 recordings with a natural sense of ambience were fairly rare, while nearly every EMI SQ recording was very natural (this includes the legendary Dark Side of the Moon in EMI SQ).

My personal gold standard is a uniform, evenly distributed 360-degree soundfield, without sound piling up towards the speakers. Ideally, the speakers and the listening room should completely disappear, with no sense of a frontal "proscenium" or stage, and no awareness of speaker location. If the speakers are low diffraction (rounded corners) and have a rapid-decay time signature (less than 1.5 mSec), that will release the sound from the speaker location, and if the decoder is not corner-focused, all of the machinery of sound reproduction will disappear. If Malcolm has decent speakers, this is what he will hear ... and best of all, the decoder can do that with stereo recordings, provided the producers didn't go too far with multimiked-mono techniques and the reverb is good quality.

Even some fairly poor SQ and QS decoders can sound quite "discrete" if they are fast and aggressive in their action, but if the underlying dynamic matrix is not well thought through, artifacts will be noticeable ... clicks, momentary harshness that comes and goes, an uneasy "swimming" sensation for the ambience, these are all unwanted artifacts that detract from quad or 5-channel reproduction. Another unpleasant artifact is the leading transient ending up with a different location than the rest of the sound ... and this is the primary challenge for any dynamic decoder.

The ideal is open space, freedom from localization artifacts, and a natural, real-world tonal quality. This is a challenge for the amps and speakers as well, and an area where most home-theater receivers fall down pretty badly. If they can't do plain old 2-channel stereo without sounding kind of nasty, then quad or 5-channel will entertaining, but not natural, and over time, fatiguing. This is the bad rap that quad and 5-channel have been fighting against since the 1970's ... unnatural sound and listening fatigue. It takes hard work to get around those problems.

 

[sjcorne]

For my own system back in the day, I always preferred the Shadow Vector decoding to a not-so-great CD-4 disc recording, but that was partly because of the low fidelity of the JVC CD-4 system, and show-offy "discrete" mixes that tended to localize sounds right at the corners ... the quad equivalent of "ping-pong" recordings.

I have the say that this is the main aspect that drew me to quad and surround music as a whole - and judging by the type of mixes that rate highly on the polls here, I don’t think I’m alone.

 

[Lynn Olson]

Interesting take on quad sound. I find that too much energy in the corners is unnatural and fatiguing, just as hard localizations of L or R are fatiguing on headphones. I want the speakers to disappear, not shout "Look At Me!", particularly if there are 4, 5, 7, or 11 of them. Put another way, the best speaker is no speaker ... I don't want coloration, I don't want peakiness, and I don't want to be aware of where the speaker is. I just want the machinery to disappear.

That's a key design aspect of Shadow Vector; if the decoder is correctly set up, separation is the same at all points of the compass. It's the opposite of CBS Full-Logic decoders, which strongly emphasize the corners. In a way, it's the Vario-Matrix of SQ decoders, but a lot faster and crisper-sounding. I'm not that familiar with the sound of the TATE DES, so I don't know how much they focus localizations into the corners.

I took the same philosophy into speaker design after the Shadow Vector project was wound down at Audionics. Low diffraction and rapid decay of the time signature can effectively make the speaker disappear, and increases the realism of speech and singing. I took it to an extreme with experimental speakers that looked like giant vitamin capsules, cylindrical with hemispheric end caps. If the room was darkened, you could walk right into them, since the sound was completely free of the speaker, and it was very difficult to tell where they really were. That experiment showed me that local energy storage (beyond 1.5 mSec) is the reason that the sound is pinned to the speaker ... it's a design defect, not a virtue. Getting the decay down to less than 1.5 mSec, or better, 1 mSec, requires control of narrowband resonances, reduction of internal cabinet modes, and low diffraction (no sharp edges on the corners of the cabinet).

I guess part off the reason for my feelings on this is that I remember the days of mono all too well. All of the sound pinned to the speaker, with almost no sense of spatial realism at all. After domestic stereo was introduced in 1958, it took a very long time for speaker manufacturers to get the idea that phantom images and overall spatial impression were desirable qualities ... the Brits were at least ten years ahead of the US manufacturers, thanks to BBC research on stereo perception. As it turned out, Audionics was the US importer for Radford, made speakers based on the Bailey transmission-line, and my Shadow Vector system at home had a stack of Radford amplifiers and a quartet of KEF 104ab speakers.

P.S. One drawback of low-energy-storage, low-diffraction speakers is the hard mono image is really tiny, only a few cm across, which sounds really weird. But that's the truth; diffraction and energy storage are undesirable speaker artifacts, and reducing these unwanted qualities results in hard-mono images that are very small. Back in the days of mono, speakers had lots and lots of energy storage and diffraction, the idea of "image quality" didn't exist, and a big, diffuse sound source would "open up" the dry mono sound and give a more spatially relaxed quality. The same speakers in stereo, of course, have very diffuse images, and depth is not well rendered.

 

[chucky3042]

Hi Lynn

Great comment in regard to "Low diffraction and rapid decay of the time signature can effectively make the speaker disappear". As you may not have been aware I have designed electrostatic load speakers since 1995 - please look up the Nakamichi Dragon electrostatic- its one of mine. I held the same theory as you for many years that in fact this resonant energy storage was indeed the defining element in transparency and as such the impulse decay response is god to me. For example a typical cone/ box speaker continues to "bong" for 2 - 3 ms after the impulse, our electrostatics achieve 80 microseconds - say 35 times faster.

Around 10 years ago "Dave the *****" and I examined the age old problem of why often a good sounding speaker tests poorly and a good testing speaker can sound bad. We tested bundles of known speakers in many configurations.......frequency response, phase response, THD. IMD, polar response and impulse decay response. In the end virtually nothing lined up (except polar in a way) that was predictive of a good sound that vanished.

Disgusted with ourselves we then had a go back to basics session (cross legged on the ground) and looked visually at things. We then tried out human trials of test jigs we made that tested the effect of "concentricity" and if the speaker was a monopole, dipole or bipole.

Concentricity: 99% of speakers have their drivers separated - woofer , mid, tweeter. We constructed jigs that would instantly switch the configuration to concentric. The result was 10 out of 10 of the audience had a major and obvious preference to the concentric. The differences was a more focused image and additional width and transparency.

Monopole, dipole or bipole: Again instant jigs were made. 9 out of 10 greatly preferred the dipole, followed by the bipole and last the monopole. Turns out that 99% of speakers made today are monopoles. The effect was the same as before way better imaging and transparency.

We then combined the both in a universal jig (still got it) and wow we created a speaker with cones that sounded 70% electrostatic. We then stupidly realised that is what we were designing for years but really did not fully grasp why the electrostatic sounded so lifelike. Think about it what instrument do you know behaves like a non concentric monopole - almost none , most are closer to a concentric dipole. I think psycho-acoustically we are adapted to interpret sound this way.

In the end our view was that the impulse decay response was the icing on the sonic cake and is what really separated the electrostatic.

I see we have walked similar roads but you were ahead of me!

Regards

Chucky

 

[Soundfield]

I’m in total agreement with Lynn and Chucky here.

I hate speakers that image poorly (and that’s probably the majority) – but it doesn’t seem to command much interest from manufactures. That may be partly because a lot of listeners don’t have an image in mind that they expect to hear accurately reproduced. Most pop and rock music is not wholly acoustic and therefore there is no real world image to faithfully reproduce - in a sense any positioning of electronic instruments, however vague is as good as any other.

Recreating an entirely acoustic performance accurately in three dimensions (or even two) is much more challenging. In my experience it all comes down to size and complexity – an excess of either is the death knell for imaging.

I have found that speakers with a narrow baffle / front aspect image well. I have always thought this was due to the minimisation of diffraction.

I have also found that speakers with multiple drivers image very badly. I assume that this is due to the absence of concentricity highlighted by Chucky that such an arrangement demands and by the decay of all that stored energy in all of those cones flapping about. Most speakers with multiple drivers also tend to have terrible vertical lobing effects such that they sound different as you sit down and stand up. I currently have speakers with three driver units closely spaced on a narrow baffle and they image very realistically. When I see speakers with six or more drivers spread out over a huge baffle I cringe!

I guess the ideal we are seeking is an approximation of a point source (or perhaps at most, the size of the original microphone capsule?). Difficult to achieve if you want to move a lot of air of course! But the tricks that you can play with the timing of the driving signals across an electrostatic panel allow you to get pretty close. I’ve often wanted to go that route but all of the ones I’ve been impressed by have been dipoles and therefore take up a lot of space.

 

[Lynn Olson]

The Ariels are an old design, dating back to 1993, but the measurements and sound are pretty decent for a simple 2-way speaker. If you compare the impulse response and waterfall to 90% of what you see in Stereophile magazine, it's clear how far the high-end industry has to go.

Part of the problem is the weird definition of "imaging" used by the review magazines. What they call "imaging" are cookie-cutter cutouts, like paper dolls hanging from a clothesline. No depth, no spatial impression of size or weight, just flat little razor-sharp cutouts. Real imaging is NOTHING like that; it's an impression of physicality, of size, of dimension, of tone, and most of all, a sense of presence and space. The most highly reviewed high-end speakers fall down really badly in these areas ... they sound unreal and very "electronic" in nature. Looking at the impulse and waterfall curves gives the game away ... lots of energy storage over a large spatial area (on the front baffle). Every time I see a $100,000 speaker with sharp prismatic edges I go "ugh, no, that's not the way to do it".

A potential "gotcha" with SQ decoders with a large number of phase shifters is energy storage starting to fall into audible regions. The EMI passive-array phase shifter SQ encoder had 6 poles, while the CBS op-amp cascade SQ encoder had 10 poles. The Shadow Vector only had 6 poles in a passive array ... it lost a bit of separation at some frequencies, but we didn't want to use an op-amp cascade, at a time when op-amps were pretty bad (this was before the 5532/5534 was introduced). This is why I think a QS decoding mode that bypasses the phase shifters entirely is a good idea ... unlike SQ, QS works fine without them, and they're only used to create a solid Center Back image. Since Center Back is subjectively unstable anyway, getting rid of phase shifters in QS decoding mode seems like an good tradeoff. Center Back has little musical merit, while the improvement in clarity by deleting the phase-shift array would be apparent on all recordings ... true, it would only be audible on speakers with very clean and fast decays, but why not?

Here's the impulse and waterfall data for the 1993-vintage Ariels, using a 6 mSec window in MLSSA and a 160 kHz sample rate. They were originally commissioned to replace a stacked pair of Quad ESL57's, and they do a fairly good job of replicating them.

 

[chucky3042]

Agree with Lynn

Imaging is a broad definition. The best test for me is close your eyes listen - then open your eyes. If there is no way in the world that the sound could possibly be coming from the speaker boxes then it probably is imaging well. Also no headaches at high volumes after long listening times.

Here is an impulse decay response of our then 600 mm long electrostatic panel:

Most of the bongs are gone after 100 microseconds and say 15 - 20 db down. The frequency scale was 100 Hz to 20 khz from memory


Oh and here is a similar response of a Martin Logan summit electrostatic that sounded GOOD but tested CRAP. Go figure???????????????????? Shows that sometimes you can stuff up an electrostatic!!

 

[chucky3042]

All the above waterfall graphs show we are all listening to a **** sandwich of lots of bonging after the impulse. Its a miracle how our brains decode this mess, I think we evolved to have the Hass response for a reason and 12 db is indeed a magic number!

 

[Lynn Olson]

The Liberty waterfall curves have an unusual linear-frequency display, if I'm reading it correctly (the major divisions all have equal spacing). This means half the display is taken up by the 10 to 20 kHz, which has little to no contribution to imaging, tone quality, or much else ... just a bit of sparkle right at the top of the range.

My own subjective weighting for audibility of time distortion is to follow the Fletcher-Munson curve, with greatest audibility in the 500 to 5 kHz region, with greatest sensitivity in the 2 to 5 kHz range. This is where speech and singing consonants fall, along with most of the harmonics of musical instruments, and is also the region of greatest sensitivity to nonlinear (harmonic and IM) distortion. The "telephone bandwidth" of 300 Hz to 3 kHz is also important ... Bell Labs found in the 1920's this was the most critical region for understanding speech over telephone lines.

My take is that less than 300 microseconds of decay clutter in the 1 to 5 kHz region is superb performance for any loudspeaker. Less than 1 millisecond is still very good, and better than most audiophile speakers on the market. Above 7 kHz, I'm not too sure much of anything is audible, except for high-Q ringing which contributes a sense of "edge" to the sound. Ringing in the 3 to 5 kHz region sounds more like "grit", and is objectionable in many high-end speakers.

The problem with measuring large ES panels is multiple arrival times from the large diaphragm; it's very difficult to tell if this is incoherent radiation (bad), ringing (worse), or simply multiple arrivals (reasonably benign). They are not the same, although they can look the same on a waterfall display. This is where I use a close-up microphone reading, about 1" or 2 cm away from the diaphragm, and "scan" the surface of the diaphragm to look for trouble areas.

We seen the same problem in speakers with a large vertical array of small drivers ... this is kind of a fad in high-end audio, but it makes the speaker really hard to measure, since at low frequencies, the drivers all radiate as one, but at high frequencies, the radiation breaks up into many small "fingers", and gets even worse in the time domain, with many arrivals coming in at different times. These kinds of speakers can have an odd incoherence and rather strange imaging, and a tonal and spatial quality that is sensitive to listening distance.

I suspect there is an interaction between type of speaker and the action of a dynamic decoder. It has to be kept in mind that CBS used big, early-Seventies JBL's for their studio auditioning (that's what they used at their 1975 demo at the Chicago Consumer Electronics Show), and these speakers had some of the worst time responses ever made ... as well as many severe peaks in the working range. EMI, by contrast, used BBC-type speakers, with much flatter responses and more coherent radiation patterns. I'm pretty sure that speakers with more coherent radiation patterns, and less energy storage, lets you hear the action of the decoder more clearly, as well as the coherence of the original mix.

As you can probably tell from this series of postings, I'm not a fan of vintage equipment, with a few exceptions (Quad, Altec, Western, Sansui, etc). But there's some good old-school technology out there, which includes hundreds of first-rate matrix-encoded records with fantastic mixdowns. It's also wonderful that many, if not most, stereo recordings expand really well into 4 and 5-channel surround-decoded versions ... if the decoder is good enough.

 

[chucky3042]

The Liberty waterfall curves have an unusual linear-frequency display, if I'm reading it correctly (the major divisions all have equal spacing). This means half the display is taken up by the 10 to 20 kHz, which has little to no contribution to imaging, tone quality, or much else ... just a bit of sparkle right at the top of the range.

My own subjective weighting for audibility of time distortion is to follow the Fletcher-Munson curve, with greatest audibility in the 500 to 5 kHz region, with greatest sensitivity in the 2 to 5 kHz range. This is where speech and singing consonants fall, along with most of the harmonics of musical instruments, and is also the region of greatest sensitivity to nonlinear (harmonic and IM) distortion. The "telephone bandwidth" of 300 Hz to 3 kHz is also important ... Bell Labs found in the 1920's this was the most critical region for understanding speech over telephone lines.

My take is that less than 300 microseconds of decay clutter in the 1 to 5 kHz region is superb performance for any loudspeaker. Less than 1 millisecond is still very good, and better than most audiophile speakers on the market. Above 7 kHz, I'm not too sure much of anything is audible, except for high-Q ringing which contributes a sense of "edge" to the sound. Ringing in the 3 to 5 kHz region sounds more like "grit", and is objectionable in many high-end speakers.

The problem with measuring large ES panels is multiple arrival times from the large diaphragm; it's very difficult to tell if this is incoherent radiation (bad), ringing (worse), or simply multiple arrivals (reasonably benign). They are not the same, although they can look the same on a waterfall display. This is where I use a close-up microphone reading, about 1" or 2 cm away from the diaphragm, and "scan" the surface of the diaphragm to look for trouble areas.

We seen the same problem in speakers with a large vertical array of small drivers ... this is kind of a fad in high-end audio, but it makes the speaker really hard to measure, since at low frequencies, the drivers all radiate as one, but at high frequencies, the radiation breaks up into many small "fingers", and gets even worse in the time domain, with many arrivals coming in at different times. These kinds of speakers can have an odd incoherence and rather strange imaging, and a tonal and spatial quality that is sensitive to listening distance.

I suspect there is an interaction between type of speaker and the action of a dynamic decoder. It has to be kept in mind that CBS used big, early-Seventies JBL's for their studio auditioning (that's what they used at their 1975 demo at the Chicago Consumer Electronics Show), and these speakers had some of the worst time responses ever made ... as well as many severe peaks in the working range. EMI, by contrast, used BBC-type speakers, with much flatter responses and more coherent radiation patterns. I'm pretty sure that speakers with more coherent radiation patterns, and less energy storage, lets you hear the action of the decoder more clearly, as well as the coherence of the original mix.

As you can probably tell from this series of postings, I'm not a fan of vintage equipment, with a few exceptions (Quad, Altec, Western, Sansui, etc). But there's some good old-school technology out there, which includes hundreds of first-rate matrix-encoded records with fantastic mixdowns. It's also wonderful that many, if not most, stereo recordings expand really well into 4 and 5-channel surround-decoded versions ... if the decoder is good enough.

Whoops
Did not notice that was a linear frequency scale. Here is one of the Nakamichi Dragon hybrid electrostatic V panel we designed.....ON A LOG SCALE!

Again the bulk of the bong is over (over 12 db down) after the first 200 microsecond time slice

 

[Lynn Olson]

Your speaker has the hallmarks of an excellent electrostat ... the long-duration, high-Q energy storage is pushed up to the 8-12 kHz region, where audibility is very low. There's moderate energy storage in the 1 to 3 kHz region, over a 1 to 1.5 mSec interval. Audible, but not a big deal. Only a few dynamic drivers can match that midrange performance ... annoyingly, the modern high-tech rigid drivers are worse, not better, than classical paper-cone and polypropylene cones.

(A close comparison shows the Ariel and Nakamich electrostat to have remarkably similar waterfall responses. A little bit of clutter in the 1-3 mSec region, and a little bit at higher frequencies. To the extent that speakers like this have any character at all, it's in the small clutter regions, which the ear/brain assigns a tonality. Most speakers have far more energy storage, which affects the clarity and naturalness of solo singers and mass chorus, as well as the tonality of instruments.)

I found that the floor bounce had to be suppressed by 20 dB (preferably more) to get decent waterfall results. In my living room, that amounted to a pile of heavy winter coats, blankets, and pillows about 0.5 meter high, right at the midpoint between the drivers and the measurement microphone. Annoying, but that's what it took to clean up the time and waterfall responses.

In terms of subjective listening, the ear automatically compensates for, and even expects, the floor bounce to be there. Since conventional floor-standing speakers are the industry norm in the final sign-off for the mastered recording, any potential coloration from the floor bounce is already equalized for in the recording process. It's the absence of floor bounce in a listening environment that would sound weird.

However ...when measuring speakers, leaving the floor bounce in the time response makes an incredible mess of the FFT transformations into frequency response curves and the waterfall display. In my experience, I found no way to mentally unscramble the mess. It was better to use absorption methods to get rid of the damn thing ... and it took a LOT of absorption to get it down by 20 dB or more. A surprising amount. Heavy carpeting only removed floor reflections above 8 kHz, and had almost no effect in the critical 1 to 3 kHz region. The only thing I found that worked was a gigantic pile of all the soft clothes and bedding in the house ... which told me a few things about the effectiveness of a 1/2" layer of felt lining inside a speaker. Polyfill was also surprisingly ineffective ... it looked nice, but didn't do much for midrange absorption.

The primary use of raw time data (impulse response) is debugging and finding the sources of unwanted reflections ... floor bounce, diffraction off cabinet edges, and the reflection off the back wall of the enclosure. All are fairly obvious if the speaker has an inherently flat response. If the frequency response is a mess, though, it's really difficult to interpret an impulse response display.

This an area where a lot of high-end speakers fall down. The frequency response is already a big mess, and the marketing department, or the magazine reviewers, want the peaks and bumps to stay in there, as part of the "house sound". That makes it really hard to debug the time response, since everything is so cluttered you can't see fine details like diffraction or internal reflections in the cabinet (although they are quite audible). So they get ignored.

P.S. On a related topic, I imagine that you found fast-decay speakers revealed the performance of dynamic decoders much more clearly than speakers with more ragged responses. Speakers with peaks and lots of energy storage tend to draw attention to themselves, and the soundfield between the speakers is vague and diffuse, even in conventional stereo playback. This is unfortunately the norm in high-end audio ... the reviewers prefer speakers with dramatic colorations, and aren't really aware of the naturalistic qualities of the soundfield ... or don't care.

 

[chucky3042]

Hi Lynn

You really do know your stuff! I cannot say that I have done much listening of surround decode with conventional decoders, all of my work was post my development of the Nakamichi's. I really have been totally spoilt since the age of 16 when I purchased my first Stax electrostatic headphones. I have only owned one set of conventional cones. Same with Dave the *****. All I can tell you is that we both get headaches within 10 minutes of listening to most cones. This tells me our brains have got lazy.

Sad story on the Nakamichi Dragons, only 200 ever built and 140 sets sold (I have 2 sets....no 1 and 2 and the golden production samples!). The parent company "The Grande group (they also own Sansui, Akai, Emerson) had a $600 Million loss and cut Naka's budget , our then stupid management refused to lower the build price by $200 each set and the product was dropped, we rebadged it to our internal brand name "Whise" - I said it would not sell and I was right. Ended up selling all 140 sets myself. 20 sets were buried in a land fill in UK somewhere I found another 40 sets in a warehouse in Holland, 20 were stolen by an employee that I recovered. In the end I lost personally 3 million. We re invented ourselves as Involve Audio!!

Grande group specialise in grabbing great names and then marketing cheap Chinese crap under those great old brand names.

Oh well - I got a couple of sets of great speakers out of it.

 

[Lynn Olson]

The arrogance. stupidity, and short-sighted greed is why I left the industry for nearly ten years ... from 1979 through 1989. I made a vow to NEVER work for another hifi company ever again, no matter what they offered me. What happened is I ended up writing for a little club magazine of the Oregon Triode Society, which David Robinson, the editor, turned into a real magazine (Positive Feedback) over the decade of the Nineties. I also wrote for Ed Dell for Glass Audio magazine, and Charley Kittleson for Vacuum Tube Valley, writing about the history of Western Electric and one of the most famous amplifiers of all time, the WE92, 91, and 86A, which all used the WE 300B, also developed in-house at Bell Labs.

The great tragedy in high fidelity is the loss of know-how and technology over time because of trade secrecy; the wheel keeps getting re-invented every few decades. There's also far too much Not Invented Here going around: the BBC pioneered the use of cumulative decay (waterfall) measurements in the mid-1960's (courtesy of D.E.L. Shorter), yet Altec and JBL didn't start using these techniques until the late 1980's, a full two decades later. For that matter, Neville Thiele's papers languished in an obscure Australian technical journal for a decade before Richard Small revised and simplified it for his doctoral dissertation, which was then picked up and re-printed by the American Audio Engineering Society in 1973. All of the speakers in the 1960's could have used modern T/S modeling, but nobody in the USA could be bothered to read anything published overseas. That's been the story of hi-fi for more than seventy years now.

That's why I've so pleased to see what Malcolm is doing in the UK and chucky3042 is doing in Australia. Now that the great labs of the past, Bell Labs in the USA and the BBC Research Labs, are in the rear-view mirror, it's up to us to get things done.

 

[quadsearcher]

I remember that VTV article! I used to drop everything and read VTV cover to cover when it arrived. I called Eric (IIRC) to talk about 7591 tubes when they were out of production. I think I even talked to Charles, kinda grumpy at the time...

 

[fizzywiggs41]

Say Lynn ,


Is there a reason why you used (for testing difficult SQ encoded tracks), the "10 cc -The Original Soundtrack " album , when you were previewing the Shadow Vector decoder back in 75 ??


Are we considering this a "closet " SQ encode ?

 

[Lynn Olson]

Hi all, my survey of surround sound is finally up at Positive Feedback Online. My warmest thanks to Malcolm Lear, chucky3042, and George Klissarov of exaSound for their pioneering work on surround sound, which I truly appreciate. And ... all the people on this forum!

P.S. No, I'm not drunk in the headline picture ... I actually hardly drink at all. It was a long, hot show at the RMAF in 2015, and I was pretty out of it when the picture was taken. The fellow on the right is Thom Mackris, my audiophile neighbor, and head of Galibier Designs. Also, check out the list of favorite surround recordings at the end of the article ... in DTS Neo:6, they sound pretty good. I imagine on the Surround Master V2, or the Shadow Vector, they sound even better.

Surround Music in the 21st Century

 

[proufo]

Thom Mackris, my audiophile neighbor, and head of Galibier Designs.

Had to think about it. Thom is from the original Teres turntable design team. Loves Ferraris (check the names of his Galibier turntables).

 

[Lynn Olson]

"10cc - The Original Soundtrack" is conventional stereo, although entertaining in DTS Neo:6, QS, or EV4 decoding, where it fills the room. This is to be expected, since a lot of the content is random-phase, and a QS or EV4 decoder will assign equal energy to every speaker when that happens. An SQ decoder will as well, but there will be no rear directionality (it takes a 90-degree phase shift between the channels to assign L/R directionality to the rear channels). By contrast, rear directionality in QS and EV4 is a function of overall L/R balance of the two input channels ... this, in fact, is the primary difference between SQ and QS/EV4 or other "regular matrix" systems.

The funny thing about Thom is he only rarely listens to CD's, and doesn't really care for digital that much. Then again, the quality of sound he gets from his turntable, arm, moving-coil cartridge, and Thoress preamp is superlative, so I can't blame him. Going out on a limb, it's pretty much the best analog I've ever heard, but then again, I'm deep into the all-tube, all-triode, all-analog sound myself, so it's a sound I appreciate.

My setup doesn't have a turntable just yet. I've got a Technics SL1200 sitting new-in-box in the basement, a retipped Supex SD900E moving-coil cartridge, and am awaiting a new preamp from John Atwood. Thom might have a dinged-up parts special where I could build my own Galibier turntable, but I don't actually have the space for the analog setup at present. As mentioned in the article, I have an exaSound e20 (ESS 9018 and solid-state) and Monarchy (Burr-Brown PCM-63K and vacuum tube) DACs for stereo, as well as a Marantz AV8003 and MM8003 for surround.

I'm due to review the exaSound e38 in the coming month, and there's a reasonable chance I'll end up buying it. On the other hand ... Malcolm Lear tells me his prototype Shadow Vector only has an ADAT output, and there are no multichannel ADAT to USB translators. Always lots of fun in the multichannel arena.

 

[Lynn Olson]

Followup: I did end up buying the exaSound e38 Mark II, and Involve Audio also sent their evaluation kit (a bare board with analog inputs and outputs, and power supply). Malcolm's Shadow Vector continues in development, and the next version of the exaSound e38 might have an ADAT input to support the Shadow Vector. (Note: ADAT is a semi-pro digital interface that uses physical Toslink transmitters, receivers, and optical fiber, but with a different protocol that supports multichannel.)

My comments thus far on the exaSound e38 Mark II are here (scroll down for my contribution):

PFO Awards

 

[Sonik Wiz]

Followup: I did end up buying the exaSound e38 Mark II, and Involve Audio also sent their evaluation kit (a bare board with analog inputs and outputs, and power supply). Malcolm's Shadow Vector continues in development, and the next version of the exaSound e38 might have an ADAT input to support the Shadow Vector. (Note: ADAT is a semi-pro digital interface that uses physical Toslink transmitters, receivers, and optical fiber, but with a different protocol that supports multichannel.)

My comments thus far on the exaSound e38 Mark II are here (scroll down for my contribution):

PFO Awards

Always glad to see a new post by Lynn here!