SQ Shadow Vector Soundfield Mapping

QuadraphonicQuad

Help Support QuadraphonicQuad:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
Thanks so much, that test record in wave or flac would be most helpful. I've just been listening to the entire decoded DSOTM and it sounds just so natural, no ping pong effects or perceivable pumping although I guess varying coefficients rather than gain eliminates that problem. I've also allowed the vectors to modify at 5-7 times per second across all 10 bands, so its very reactive to primary sound sources.
 
There still seems to be a very low level noise issue from the vector filters so maybe good quality classical tracks could help track down the problem.
maybe one of the Masterworks SQ LPs that contain Quad material Vocalion have reissued on Surround SACD?
something like the Pulcinella Suite could be useful to have the discrete masters to compare to the decodes?
pretty sure i have the SQ LP of that one.

Not to keeping plugging my analysis work but the Bartok Concerto SQ LP is another good choice- the cover gives you clear identification of instrument placement and we have the Q8 to compare against.
 
Last edited:
Oh, OK, thanks for letting me know - I was too slow off the mark then!
What about the Bartok Concerto, have you sent him that as well?

noo problemoo! ;)
i only have that Bartok Concerto on MCh SACD (which sjcorne has established is not the Quad).

tbh i have v.few Classical Quads & 5.1's compared to other genres etc., i'm a pop & rock pleb but i'll try to help out as much as i can! :p
 
Hi all!!! I want to thank Malcolm Lear for reviving the Shadow Vector; it's been decades since I last heard it up and running. I never did figure out how the Tate DES did it's thing, but reading between the lines it seemed similar in decoding action to the CBS Paramatrix (except faster) and favored the six SQ cardinal points. Shadow Vector has symmetric 360-degree decoding at all intermediate localizations, and can be calibrated with a pair of oscillators that are slightly different in frequency (look for smooth cancellations in the "off" channels as the sound sweeps from LF to CF to RF to CB.

In subjective terms the SV sounded noticeably more spacious than the CBS Paramatrix and TATE DES units, but this is a 45-year-old memory on a set of KEF 104's in what later became the EBU speaker layout. Compared to a discrete master tape it actually sounded more spacious on direct switchover, and that could be considered a subtle artifact of the SV matrix not seeking cardinal points unless directly encoded as signal sources. (A design goal was symmetric 360-degree presentation of the random-phase reverb content present on many recordings.)

I sent a private PM (or whatever it's called on this forum) suggesting to Malcolm Lear that the unit he's working on has EV4 as well as SQ and QS decoding. EV4 avoids the potential sonic degradation of allpass phase shifters, while retaining separation in the rear channels. I also personally think dynamic EV4 decoding is optimal for decoding good stereo records, and would certainly prefer it to Dolby Pro-Logic II (music mode) or DTS Neo:6 with stereo sources.

Once again, congratulations to Malcolm Lear for what can only be called a labor of love. Yeah, I'm a little sad that Quad went nowhere in the mid-to-late Seventies, and more annoyingly, was intentionally cut out of the picture during the standards-setting for the Compact Disc. And I have to agree with others that a Center speaker doesn't do much for music presentation, with an odd flattening of the central soundstage (in depth, on the Z-axis).
 
Something that's only mentioned in passing in the Shadow Vector Patent (#4,018,992) is the dynamic decoder has exactly the same treatment for random-phase reverberation as a simple static decoder. The underlying design premise of the SV is that reverberation in most recordings (whether from an EMT plate or spaced microphones) is random-phase. In visual form, that means it uniformly covers the entire surface of a Scheiber sphere, while localized instruments look like pinpoints of light that are rapidly modulated by musical transients.

The task of the dynamic decoder is to mimic human spatial perception by localizing, then tracking these intermittent sources, while the overall spatial impression is stable, diffuse and non-localized. Localization and tracking of the instruments are important, but that should not impair the spatial impression. In fact, if the spatial impression is moving around, that's very distracting, since that never happens in nature (unless you like to jump around as you listen). But normally, the spatial impression is diffuse and stable in location, while the instruments pop up and recede. A correctly designed dynamic decoder will have variable resolution for non-enhanced locations (reverb), while transients are crisp. (The attack times of the prototype were 1~3 mSec and release around 20~50 mSec, following the perceptual timings of the Dolby A and B systems. In a multi-band system like Malcolm Lear is doing, the attack and release times can be optimized for each frequency band.)

This is why the Shadow Vector does not favor the cardinal points, which I think is different than the TATE decoder, and maybe Dolby Pro-Logic II and DTS Neo:6. I hear what I call "detenting" with these decoders ... the reverberant impression sort of piles up around the speakers, instead of filling the room. That sounds very unnatural to me, and sort of defeats the whole point of quadraphonic or surround, which is to break down the walls of the room. If your attention is drawn to the speakers, the quad system isn't doing its job correctly. The speakers should disappear completely, which is why transitioning to speaker design later in the Seventies wasn't as big a jump as I thought it would be.

Some of the criticism I've attracted over the years for being a proponent of dematrixing stereo mixes is from people who have never heard a correctly engineered dynamic decoder ... and I would be the first to admit that a mediocre dematrixing of a generic stereo mix is pretty unnatural sounding, and sometimes downright distracting and annoying. "Slow" dynamic decoders fall in this category as well, since it can sound like the whole soundstage is sloshing around as the decoder is too slow to follow the musical line. I get that. But the solution is a better decoder ... quick in action, and fully symmetric in 360 degrees.
 
I couldn't agree more Lynn - I have always thought that the way that a quadraphonic system handles environmental / ambient sound is a key measure of its performance. It is ironic that many quad systems are actually less good in this regard than a stereo recording made with a simple co-incident pair of microphones and played back over decent speakers! So many quad recording techniques and hard logic decoders as you say, artificially highlight a limited number of cardinal positions - it might sound initially impressive but realistic in purely hifi terms it certainly isn't. I hope Malcom Lear, through his hard work, is able to bring his SV implementation to a saleable form so that more of us could enjoy your vision after all these years!
 
...That sounds very unnatural to me, and sort of defeats the whole point of quadraphonic or surround, which is to break down the walls of the room. If your attention is drawn to the speakers, the quad system isn't doing its job correctly. The speakers should disappear completely, which is why transitioning to speaker design later in the Seventies wasn't as big a jump as I thought it would be.

I always thought this is more the fault of the quad mixes themselves rather than the decoders. CBS quad mixes are generally very dry-sounding and often have instruments isolated in a single speaker, so naturally the attention is drawn to the speakers.
 
I agree, closely miked individual instruments recorded in an largely dead studio, pan-potted into artificial cardinal locations obviously don't convey much sense of acoustic space regardless of the decoder. A slightly tougher decoder test is how stable and all enveloping are the decaying reverberations of, say, a pipe organ in a cathedral several seconds after it has finished sounding.
 
Wow, Lynn's visualisation of the Scheiber sphere of incoherent reverberation covering the surface with localised hot spots really hits the nail on the head. Whilst developing the software implementation you could easily visualise the 4 channel vectors spinning round into position to best decode the hot spots without any loss of information. I have found the decodes to be very immersive and quite often preferable to discreet. The comments on the decoders dynamic rate of change is so very true. I spent a great deal of time determining the best settings and certainty the luxury of multiple bands helped a great deal, since it was possible within limits to increase speeds on the higher frequency bands. However the biggest advantage of a digital solution is lookahead. To start changing the decoding ahead of an event and finish it just after in a balanced way really helped in stabilizing image location and allowed slightly slower control rates which eliminated the odd artifacts caused by bad encoding. Definately worth all the effort and hopefully we can get a complete prototype very soon.
 
The electronics of the original CBS encoder (the standard 4 -> 2 encoder) are kind of marginal. The SQ matrix relies on the precision of the all-pass phase shifters for accurate encoding and decoding. Put another way, the more poles of phase shift, the higher the separation in playback. One of the less happy thing about the SQ matrix is the use of 8 poles for encoding, and 6 poles for playback. The phase shift is cumulative, so that's a net 14 poles of phase rotation across the audio band ... and a conventional 2 or 3-way loudspeaker adds a 2 to 6 more poles on top of the SQ encode/decode process. (It takes a digital FIR filter to unwrap the phase.)

On top of the potential audibility of that much phase shift, the CBS implementation used a cascade of 301A opamps, which are pretty marginal from a modern viewpoint (very low slew rate which results in HF distortion), and these opamps are coupled by 100uF electrolytic capacitors (because they have significant DC offsets). The early opamps, such as the 709, 741 and 301, are no longer considered acceptable for audio except in the very lowest-quality applications. The first decent opamp (which was specifically designed for audio) was the Philips/Signetics 5532 (single) and 5534 (dual), and these didn't come out until 1978 or so, seven years after the CBS professional encoder.

EMI took a different approach for their encoder. I saw and understood the schematic of the CBS professional encoder, but when I visited EMI Labs in person in 1975, they told me the EMI phase shifter was a passive parallel matrix with discrete transistor electronics. (I'm guessing with 6 poles, since 8 would require extremely precise capacitors (all of the poles interact in a passive parallel matrix). A cascade phase shifter like the CBS approach is much easier to build and tune, but then you have a long cascade of electronics.

One of the big differences between SQ and QS encoding is that phase shifting is optional for QS (you only lose the Center-Back localization), while it is mandatory for SQ (the 90-degree phase shift encodes Left Back and Right Back). The EV4 system does not use phase shifters on either encode or decode sides, and is pretty close to the Sansui QS system (just slightly different encode points).
 
Last edited:
One of the odd properties of the SQ matrix are the two diagonal splits, and the 4-channel sum (or center-top) being prohibited. QS and EV4 are the opposite; there are no diagonal splits, but 4-channel sum (or center-top) encoding is easy: simply L at zero degrees while R is delayed 90 degrees (the same as Left Back in SQ).

What's a bit more surprising is the phase relationship of the output channels of the SQ decoder; one of the diagonal splits is in-phase, while the other one is out of phase. This can be corrected by inverting the phase of one of the rear channels (I forget which one), but then Center Back is out of phase. I suggested back when I was Audionics giving the user the choice of either phase selection ... standard SQ playback with Center Back intact but asymmetric diagonal splits, or out-of-phase Center Back but symmetrical diagonal splits.

I personally think Center Back is a disgusting localization, only good for horror-movie effects, so losing it is no big loss. It's also hard to localize, even in a fully discrete mix, since it tends to flip overhead and appear in front. Try it for yourself: listen to ordinary 2-speaker stereo while facing directly away from the speakers, and notice how difficult it is to localize anything. The phantom center will be very vague and diffuse, and also quite fatiguing and unnatural-sounding, with a noticeable hollow coloration. That's why this localization is not really suited for musical use; it's more a gimmick for quick pans across the rear.
 
39533
 
As you can see from the above images, there is no location on the Scheiber sphere that corresponds to a 4-channel sum. A diagonal split, though, is simply the path between Left Front and Right Back, and another path between Right Front and Left Back. Since the Shadow Vector is agnostic about encode location, they play back with full separation.

I should also note the standard CBS 4 -> 2 encoder has a deviation from ideal as the signal is panned across the sides of the room; I think the pan-locus bends towards the front for the Left Front -> Left Back path, and towards the rear for the Right Front to Right Back path. It was possible to skew the logic control to offset this, but I had to draw the line somewhere ... besides, there was no way to tell which of several possible encodes that CBS was using ... the standard 4 -> 2 unit, the one with selectable front or rear bias, or the Position Encoder.

The biggest advance with an all-digital decoder is, as mentioned by our intrepid Malcolm Lear, is getting rid of annoying transient clicks caused by the finite delay of the logic circuit, which in the analog version is always slightly behind the audio signal by a few milliseconds. The clicks are subtle, appearing only in the suppressed channels with program transients, but better to sync up the audio and control systems. The implementation of logic sensing is also much better, since you don't have to wrestle with fast-acting AGC circuits in the analog domain, and you can do little tricks like velocity sensing (program transients control the speed of the logic circuit).

P.S. Funny side note: these images originally appeared to me as animations as a I came down from 150 micrograms of LSD. They were so vivid (imagine them in dark space with the light chasing around the sphere, and all the vectors trying to hide from the light) that I had to write them down immediately. A few weeks later I ground through some of the math, realized it was indeed possible to built the thing, and spent about a month down at the Los Angeles Patent Library. They told me about filing a Document of Disclosure, which formed the basis of the patent, complete with these drawings, which summarize the action of the Shadow Vector.

P.P.S. Once the prototype was up and running, and being demonstrated at the Consumer Electronics Show in 1974, the CBS Labs design team came over and visited the Audionics suite. I was surprised to find they didn't use Scheiber notation, which would make describing a dynamic decoder incredibly complex and opaque. I was even more surprised that they actually hired an outside mathematician to describe all the weird vectors used in the CBS Paramatrix, while it's intuitively obvious how the Shadow Vector works, thanks to the Scheiber notation. I still don't understand the complex vector math of the TATE decoder designed by Wesley Ruggles, but reading between the lines, it, like the Paramatrix, appeared to be optimized for the six cardinal points of SQ, while the Shadow Vector is agnostic about the encoding localization and linear in its action ... which I describe as "symmetric" in the patent.

It's not clear to me how Dolby Pro-Logic II (music mode) and DTS Neo:6 decoders work. They don't sound as symmetric as the Shadow Vector, nor as quick, and are much less spacious. DPL-II in particular sounds very shut-in and pretty slow compared to what I expect from good dynamic decoding, and DTS Neo:6, while sounding more open and pretty fast, can sound kind of spitty and harsh on some material. Neither sounds anything like Shadow Vector, which is kind of disappointing, considering the original insight of those moving lights goes back to 1973 or so.
 
Back
Top