I second Doug & Jefe's comments. Since joining QQ earlier this month, I am in awe of the technical knowledge of Disclord and so many others here. I thought I knew a lot about quad & multichannel. Perhaps I should call myself Linda Schultz, because I feel I know NOTHING compared to many of you. Keep those technical posts coming.
Linda
Queen of Quadriphony
Thank you all for the compliments - they make me blush because I feel that I'm only a talented amateur in what I know. I wish I was an E.E. so I could understand schematics and the actual 'nitty gritty' operation of stuff like VCA's much better - I frustrated Martin Willcocks to no end in having him help me get up to speed on vector spaces and matrix algebra - which I still don't understand at all - I understand 'just enough' to 'figure out' what's going on inside a decoder, etc... And I'm dyslexic when it comes to numbers so that can frustrate me to no end - all though high school I never understood why simple math was such a struggle for me and it wasn't until I was in my mid-20's that I was diagnosed as being dyslexic - but only with numbers. Letters are no problem - but even dialing a phone can cause me problems. So I have to read and re-read anything I write that has numbers in it.
I know as much as I do about quad simply because I have time on my hands - when I was 17 a drunk lawyer rear-ended my car, seriously damaging my back - I had many surgeries over the years and all that 'down time' gave me 'study time' to figure out quad stuff. And I had to 'live it' as it happened in the 70's - what I mean is, I knew nothing about quad in the late 70's/early 80's. Then, a mail-order firm we bought LaserDisc's from, Starship Audio Industries, was selling off Fosgate's remaining stock of the Fosgate Tate II 101A's for $249, so my dad bought me one. I didn't know what "logic steering" was, or what "SQ" or even the "Tate DES" was; I just wanted surround sound for my LaserDisc's like they had in the movies. So, I had the Fosgate and was really happy with it when, about a year or so later, the first Dolby-branded surround decoder, the Surround Sound Incorporated SSI-360 made its appearance on the scene. Since I didn't know about logic, I had no idea that the SSI unit was a simple non-logic device - it had the Dolby MP Matrix logo on the front and that's all that mattered - it was an 'official' licensed Dolby decoder, just like the theaters used! (or so I thought!) So I lusted after that, but it was over $400 at the time so I couldn't buy it. Then Sanyo came out with a stand-alone Dolby Surround decoder that not only had that all-important "Dolby Surround" logo, it also had a built-in amp for the rear channels and a cool looking green and red visual display that 'grew' as the front or back channels predominated. And all the articles in Video, Video Review and Stereo Review stated that the 20 millisecond time delay in 'official' Dolby Surround decoders was mandatory for best performance, so when the Sanyo went on sale for $179 at Silo Electronics in Albuquerque, I jumped and bought it. Almost from the start I realized something was 'different' between the Fosgate 101A and the Sanyo - when a sound appeared in Left Front it also seemed to appear at full level from both surround channels - so I kept lowering the level of the surrounds (which had never been a problem with the Fosgate!)
Then, one weekend my dad and I went down to the library's main branch in downtown Albuquerque - Parking was free on Saturdays, so that's when we went. I discovered that the library had Radio Electronics, Stereo Review and High-Fidelity magazine on microfiche going back to the 1950's - somehow, I stumbled across an article called "Matrix Quad - How Good Is It?" - it talked about SQ and QS as well as CD-4 and Peter Scheiber's initial demonstrations. Going through the library's LP collection I found a LOT of quadraphonic records, mostly in the CBS SQ format - so I had to have an SQ decoder (yes, I'm an idiot - I didn't realize that the "SQ" position on the Fosgate 101A was the same "SQ" system as the records or the articles were talking about). So, I scoured second-hand shops and the Goodwill U-Fix It Corral and I bought (for 5 bucks) a Radio Shack SQ decoder - it even had the SQ logo on the front. Got it home, played some SQ records through it and was totally disappointed. Back to the library to read more - I discovered I needed an SQ decoder with "Front-Back Logic" - well, what do you know, the Goodwill U-Fix It Corral had a Pioneer receiver with Front-Back Logic SQ. More disappointment again! Hey, maybe this SQ format really sucked and that's why it's not around any longer??? Back to the library where I found an article by Len Feldman about the QS Vario-Matrix - and what do you know, the Audio Clinic, a local electronics repair shop that sells off equipment that people don't pick up or refuse to pay repairs on - they had a Sansui QRX-6500 - and it was a Vario-Matrix decoder - the first Vario-Matrix decoder on the market in fact. I didn't know it was only a "Type B" decoder - but it did have a front-back Vario-Matrix position for SQ called "Phase Matrix" - and I was shocked at how well it decoded SQ as compared to the Pioneer receiver's Front-Back SQ Logic. The Sansui totally kept Center Front vocalists out of the rear channels. QS decoding performance wasn't so hot though - although I could hear that there were 4 channels being decoded from the few QS records I had borrowed from the library, there just didn't seem to be enough separation between each channel, other than CF to CB and vise-versa. Again, the library's Microfiche came to the rescue with an article (again by Len Feldman) about SQ Logic and it talked about "Wave-Matching" logic being combined with the Front-Back Logic to create what was called a "Full Logic" SQ decoder. There was a picture of the Sony SQD-2020 decoder (but with an awesome looking wooden case around it like my Sansui receiver had - was that wood-wrapped Sony SQD-2020 ever sold that way here in America?) - and what do you know, The Audio Clinic had a Sony SQD-2020 for $50 - they had replaced all the caps and resistors with high-quality versions and even retrofitted the SQ phase-shifters with 1% precision components (the measurement sheet that came with it showed that the phase shifters went from a tolerance of 10 degrees to only 2-3 degrees over most of the bandwidth) - that's when I also found out that the SQE-2000 was originally planned to be sold as a 'companion piece' to the 2020 decoder, but that CBS had talked Sony out of doing that and letting them (CBS) market the encoder directly to radio stations instead. The technician at The Audio Clinic had the letter to dealers outlining the change in marketing in his files and gave me copies - which, I've sadly since lost!. The other changes made to the unit were to take off the (CBS recommended) blend resistors between the front L/R and Back L/R channels - in their place they put variable resistors that lead to pots protruding from the back where they had drilled two holes. Another pot was added that worked somewhat like a "dimension" control - as you turn it down the 2020 goes from full logic decoding to basic-matrix, but it keeps Front-Back Logic at full force at all times. Apparently, the mod's were done for some rabid quad fan who had died and his wife told them to keep the unit. They had had it sitting on the shelf for so long the price just kept going down - until I came along and gladly took it off their hands. They threw in the original instruction manual and the full service manual for it too.
So, anyway, there I was, happy with my turbo-charged SQD-2020 - especially the awesome VU meters - when I find another article at the library about SQ logic, this time it was introducing the CBS Vari-Blend to be used in place of the front-back gain-riding logic circuit. And what do you know, The Audio Clinic has an Lafayette SQ-W for $75 bucks. It turned out to be defective - apparently the parts had aged causing the logic action to lose calibration, but since The Audio Clinic gave all their used gear a 30 day warranty, they replaced the leaky caps and stuff and re-calibrated the decoder. I was blown away by the SQ-W - with the SQD-2020 I could often hear vocals quickly move to the rear as a rear source predominated, then snap back to center front. It was quite annoying, especially with records that were worn or had sibilance. They seemed to cause the 2020 real problems keeping Center Front out of Center Back. The SQ-W had no such problems - once in a while I could hear the rear channels blend to mono and then 'un-blend' back to stereo, but it wasn't nearly as annoying as the sibilance.
FINALLY, I came across a 1983 review, in Stereo Review, of the Fosgate Tate II 101A - and it "clicked" for me that this Fosgate that I'd had all along, was the absolute ultimate for SQ decoding. Since I had the Sansui QRX-6500, everything tapped into it nicely for all my quad listening. I got a Panasonic SE-405 CD-4 decoder, but never had a good enough stylus to get decent separation between front and back. Later I was given the Technics SH-400 by QQ Forum member R. Scott Varner, and he gave me the 7-inch CD-4 set-up record for it too, but again, I've never had a good enough stylus to take advantage of it. (
it's also thanks to R. Scott Varner that I have the pristine CD-4 copies of "Earthquake", "Jaws" and "Poseidon Adventure", plus the instructions for the prototype CBS Paramatrix decoder and the instruction manual for the Sony SQE-2000 Encoder - oh, and the service manual for the Technics SH-400 that included a letter from Panasonic to the demodulators owner about bypassing the RIAA stage of the Quadracast Systems CD-4 IC. They said it couldn't really be done)
While I was doing all of the above, I was also going down to the University of New Mexico because they had a patent depository there - I'd spend hours there copying quadraphonic related patents - at $1 per page! The library only charged 25 cents per page for their magazine copies!
So, as you can see, I 're-lived" the whole quad era in a compressed time frame - from non-logic to F/B, to wave-matching to Vario-Matrix, to Wave-Matching with Vari-Blend and finally back around to the Tate DES... with CD-4 thrown in for good measure. My parents thought I was insane, and I think they got tired of hearing "this will be the final decoder I'm going to buy since it is the ultimate in decoding SQ."
In 1987 I got a Fosgate DSM-3606 (if the 101A was good, well then the later Fosgate's, designed specifically for movies, and co-designed by Peter Scheiber, the inventor of matrixing himself, should have been flat-out an amazing sounding decoders for movies. Boy, was I wrong and disappointed. The DSM-3606 simply was not a good decoder - it didn't use any signal cancellation between the L & R front channels and the surround channels and the front soundstage expanded and contracted like a crazy accordion whenever dialog and stereo music or effects were present at the same time. And because high quality four-quadrant analog multipliers are expensive -
the number of multipliers or VCA's a crosstalk cancelling decoder needs is the product of the number of primary directions to be sensed and the number of crosstalk bearing channels for every signal to be cancelled, so a typical, and simple, four channel quadraphonic decoder needs at least 8 multipliers or VCA's. Both the National Semiconductor "Tate-I", and Exar Systems "Tate-II", DES integrated circuits used 16, high-precision, four-quadrant analog multipliers: 8 in the 'front' matrix multiplier chip and 8 in the 'rear' matrix multiplier chip, both of which were fed from the Tate directional detector chip and the custom designed Direction Control Interface & in the Tate II design, the Automatic Dimension Control (which were the part of the original Tate IC's that National screwed up during design and manufacture) - the FET's Fosgate was using for the VCA circuitry were low cost and quite noisy, plus they distorted and overloaded easily - depending on the program it could sound as if someone was whistling behind the main soundtrack. The widescreen LaserDisc release of 'The Hindenburg' gave it absolute fits to decode correctly. The background noise level could be so high as the logic action panned sounds across the front soundstage that sounded like the movie had been recorded on a micro-cassette recorder with no form of noise reduction - while other Fosgate decoders might have performed much better - and I have to believe that because of the number of people who love their Fosgates, my experience has been overwhelmingly negative. I always wanted to audition the mis-named "six-axis" decoder in my own system - it was the only Fosgate decoder to have an encoder built in so that AC-3 and DTS titles could have all 6 channels down mixed to 2 channels - apparently, the encoding matrix wasn't too far off from Pro-Logic II, Circle Surround or Lexicon's Logic-7, which is understandable considering all of them use 'real' coefficients and don't use phase encoding. There's not to many ways to make a matrix encoder if phase encoding isn't used. I think the differences mainly lie in the amount of left or right bias applied to the out-of-phase surround signal to encode the L or R surround.
I still think the surround world would be a better place if we went back to encoders and decoders that used complex coefficents - i.e. phase encoding and decoding.
I must apologize for a reply that's much, much longer than the nice notes that preceded it.