Quantcast

(1972) Matrix encoded CD-4 (proposal)

Help Support QuadraphonicQuad:

tom3

New member
Joined
Jun 4, 2014
Messages
9
And today FM could easily broadcast discrete quad with their HD signals. Digital signals to an equipped receiver on the hybrid signal, could be interesting.
 

Marcsten

400 Club - QQ All-Star
Since 2002/2003
Joined
Sep 24, 2002
Messages
403
Location
Seattle
Someone said that if the Tate system had been available at the launch of Quad or at least early, SQ would have been the clear winner. I agree.

Would a single system had been enough to make Quad survive?
Well as I just posted elsewhere, while the confusion of multiple formats was a problem, the real problem is that NONE of the worked worth a crap! If Tates were available in 1973 SQ would have done well. If Vario Matrix had been available at launch, QS might have done well, and if CD-4 had ever worked consistently with sophisticated demodulators and cartridges that properly tracked the records that didn't cost a fortune, then it might have done well, if it was affordable. But none of them were and people loved Q8s until they jammed and the tape wrapped around the flywheel...
 

atrocity

1K Club - QQ Shooting Star
Since 2002/2003
Joined
Jun 14, 2002
Messages
1,668
Location
Sacramento, California
Dolby Surround was 4.1 - left, right, dialog, surround, (subwoofer).

Some decoders left off the dialog channel.
When "Dolby Surround" first came into the home, it was Left, Right and mono Surround.

I think some of the first home decoders didn't even remove the surround information from the front speakers, though 30+ years later I can no longer remember for sure just how I wired up that cheap Radio Shack box. :)

It wasn't until "Dolby Surround Pro-Logic" that we got a dedicated center speaker. I don't think a dedicated .1 was ever part of the spec until discrete Dolby Digital.
 

Marcsten

400 Club - QQ All-Star
Since 2002/2003
Joined
Sep 24, 2002
Messages
403
Location
Seattle
When "Dolby Surround" first came into the home, it was Left, Right and mono Surround.

I think some of the first home decoders didn't even remove the surround information from the front speakers, though 30+ years later I can no longer remember for sure just how I wired up that cheap Radio Shack box. :)

It wasn't until "Dolby Surround Pro-Logic" that we got a dedicated center speaker. I don't think a dedicated .1 was ever part of the spec until discrete Dolby Digital.
My recollection, which could certainly be wrong, was that when the Pro Logic Dolby came along they had licenses the tate or similar SQ decoders, but got left, right, and center as three of the four channels and both rears for the same final channel. No .1 as there were no channels left. Anyone else recalL?
 

Q-Eight

2K Club - QQ Super Nova
Since 2002/2003
Joined
Sep 30, 2003
Messages
3,204
Location
Castlegar, BC, Canada
Well as I just posted elsewhere, while the confusion of multiple formats was a problem, the real problem is that NONE of the worked worth a crap! If Tates were available in 1973 SQ would have done well. If Vario Matrix had been available at launch, QS might have done well, and if CD-4 had ever worked consistently with sophisticated demodulators and cartridges that properly tracked the records that didn't cost a fortune, then it might have done well, if it was affordable. But none of them were and people loved Q8s until they jammed and the tape wrapped around the flywheel...
Just my 2 cents on the topic of CD-4.... I think fundamentally, the system worked quite well. But I'm speaking about experience with my Technics SH-400 and a modern linear contact stylus and cartridge. BUT, my point is with my setup, I have a visual gauge of carrier level on any given LP. I have to adjust the Carrier Level pots on my demodulator from time to time because the carrier level varies from disc to disc. So, this has lead me to believe that perhaps it was the pressings and/or the CD-4 mastering that really needed to conform to specifications. Maybe this isn't such a problem with automatic carrier level sensing demodulators, but how sensitive and how quick are they to react and achieve carrier lock? I think CD-4 was the way to go, but JVC should've pulled a Sony and made other manufacturers stick to the plan. Cheaping out on the demodulators and stylii and cartridges is what caused most of the problems. I've been through several myself and nothing has ever given me the performance of my current setup. I even tried a Panasonic Demodulator with my current cartridge/stylus and turntable, and for whatever reason, it was a splattery mess.

Quad probably should've stayed a tape-only media until better disc systems were ready.
 

MidiMagic

400 Club - QQ All-Star
Joined
Jul 5, 2010
Messages
422
The base matrix for UD4 was the BMX matrix (often called UMX).

QMX was UMX with 4 separate channels of transmission and complementary versions of the BMX 2-channel version for the other 2 channels. QMX would reproduce a discrete reproduction of the 4 original channels when properly decoded.

UD4 was a version of QMX with the carrier channels restricted to 3KHz bandwidth.

This is all covered in the article:

"Discrete-Matrix Multichannel Stereo" D.H. Cooper & T. Shiga,
Journal of the Audio Engineering Society 06/1972 V 20 pp. 346-360

It does require a bit of mental translation since Denon changed the product names in the interim.

These three articles show the encoding of all matrix systems:

"Analyzing Phase-Amplitude Matrices" Peter Scheiber,
Journal of the Audio Engineering Society 11/1971 V 19 pp. 835-839

"4-2-4 Matrix Systems: Standards, Practice, and Interchangeability " John Eargle,
Journal of the Audio Engineering Society 12/1971 V 20 pp. 809-835

"A Geometric Model for Two Channel Four Speaker Matrix Stereo Systems" Michael Gerzon, Journal of the Audio Engineering Society 03/1975 V 23 pp. 98-106

Notice that QS, EV, DQ, and Scheiber are all minor variations on the same matrix encodings (with the general characteristic called RM).

Notice that RM, BMX, H, and Ambisonics are four different matrix systems.

As long as CD-4 requires the equivalent of a clean-room environment to work properly, I do not consider it a viable system for general public use.

Note that later CD-4 records were made with a variable level carrier. The carrier level decreased when the audio level increased so the record was not overcut. This could cause Q-eight's problem.

Radio stations hated CD-4 because it could not be slip-cued - the blank groove is not blank between tracks.

Tape as the only format? Not for longevity:

- I have very few records (out of 1000s) that have failed (the oldest is from 1908).
- - - I have a couple of World-War-II-era records that delaminated and fell apart because substitute materials were used to make records due to rationing (the cores were corrugated cardboard).
- - - All of my CD-4 records were ruined when I got them (I was a college student at the time and could buy only used records - I was given the cartridge and demodulator).
- - - For some reason only '70s RCA records developed mold on the labels.
- - - a few polystyrene records were played with a stylus force too high for styrene.

- Most of the tapes I owned in the Quad era have failed.
- - - Many of them had the oxide fall off the tape.
- - - Others had the adhesive holding the oxide glue the layers of tape together.
- - - The rubberized nonslip backings deteriorated and stuck to the tape in places.
- - - Many of the beginnings of reel tapes broke at the end of rewind and snarled.
- - - The splices in some 8-tracks and cassette leaders failed.

- The VHS tapes failed in similar ways.

- Many of the floppy discs I had in the 1980s-90s have failed in the same ways.

- I have had very few CD and DVD failures.
- - - A very few developed air leaks that corroded the foil sheets.

- I have had a small portion of my CD-R, DVD+R, and DVD-R recordings fail:
- - - Some early CD players could not play certain brands of CD-R.
- - - Several early CD-Rs were destroyed in hot cars.
- - - Some newer DVD players won't play the DVD-R discs.

- Many of the audio computer files I made in the 1990s were lost because Microsoft deprecated the formats and didn't tell anyone.
 
Last edited:

Doug G.

1K Club - QQ Shooting Star
Joined
Jul 8, 2007
Messages
1,646
Location
Rochester, Minnesota
....As long as CD-4 requires the equivalent of a clean-room environment to work properly, I do not consider it a viable system for general public use...
I wish you would stop repeating this false assertion. My listening room is anything BUT a clean room and I listen to CD-4, successfully, all the time.

Doug
 

Soundfield

900 Club - QQ All-Star
Joined
Dec 29, 2011
Messages
935
Location
Essex, UK
I would have thought that the very last thing CD4 needed was matrix processing to be added to the huge chain of processes its poor input signals were subjected to!
 

kfbkfb

300 Club - QQ All-Star
QQ Supporter
Since 2002/2003
Joined
Feb 20, 2003
Messages
371
Location
Midwest USA
Yes, as with all Matrix systems, there'll be some signals that cancel out:
^^^
"...recorded cleanly in discrete quadraphony...when this clever little "nasty"
was fed into the SQ encoder something strange happened: nearly all of
the voices slowly faded away into a soft background sputtering..."


(CD-4 decoding of a hybrid Matrix/CD-4 disc would still result
in discrete Quad)


Kirk Bayne
 

jaybird100

1K Club - QQ Shooting Star
QQ Supporter
Joined
Apr 27, 2006
Messages
1,263
Location
Pembroke Pines, FL
Someone said that if the Tate system had been available at the launch of Quad or at least early, SQ would have been the clear winner. I agree.

Would a single system had been enough to make Quad survive?
That would be a pipe dream. Maybe it would, but I'm not sure it would have been SQ. One has to take into consideration how the system would affect stereo and mono reproduction, in situations when decoding was not possible. SQ was notoriously bad when it came to stereo and mono compatibility; QS, while not perfect, was better for that. Had a Tate SQ decoder been available in the 70's, and it had to seriously compete with QS Vario-Matrix, which was available, and if the "big name" of CBS hadn't been involved, QS might have done it. In fact, if it hadn't been for Brad Miller's interference, the Warner Music Group, (WEA, in the 70's) would have gone with the QS system. This was documented in BILLBOARD Magazine back then. If WEA had joined ABC Records as a purveyor of QS, that system might have taken the lead, and perhaps the industry.
 

Wurly1

Active Member
Joined
Dec 15, 2019
Messages
56
Location
Canada
The base matrix for UD4 was the BMX matrix (often called UMX).

QMX was UMX with 4 separate channels of transmission and complementary versions of the BMX 2-channel version for the other 2 channels. QMX would reproduce a discrete reproduction of the 4 original channels when properly decoded.

UD4 was a version of QMX with the carrier channels restricted to 3KHz bandwidth.

This is all covered in the article:

"Discrete-Matrix Multichannel Stereo" D.H. Cooper & T. Shiga,
Journal of the Audio Engineering Society 06/1972 V 20 pp. 346-360

It does require a bit of mental translation since Denon changed the product names in the interim.

These three articles show the encoding of all matrix systems:

"Analyzing Phase-Amplitude Matrices" Peter Scheiber,
Journal of the Audio Engineering Society 11/1971 V 19 pp. 835-839

"4-2-4 Matrix Systems: Standards, Practice, and Interchangeability " John Eargle,
Journal of the Audio Engineering Society 12/1971 V 20 pp. 809-835

"A Geometric Model for Two Channel Four Speaker Matrix Stereo Systems" Michael Gerzon, Journal of the Audio Engineering Society 03/1975 V 23 pp. 98-106

Notice that QS, EV, DQ, and Scheiber are all minor variations on the same matrix encodings (with the general characteristic called RM).

Notice that RM, BMX, H, and Ambisonics are four different matrix systems.

As long as CD-4 requires the equivalent of a clean-room environment to work properly, I do not consider it a viable system for general public use.

Note that later CD-4 records were made with a variable level carrier. The carrier level decreased when the audio level increased so the record was not overcut. This could cause Q-eight's problem.

Radio stations hated CD-4 because it could not be slip-cued - the blank groove is not blank between tracks.

Tape as the only format? Not for longevity:

- I have very few records (out of 1000s) that have failed (the oldest is from 1908).
- - - I have a couple of World-War-II-era records that delaminated and fell apart because substitute materials were used to make records due to rationing (the cores were corrugated cardboard).
- - - All of my CD-4 records were ruined when I got them (I was a college student at the time and could buy only used records - I was given the cartridge and demodulator).
- - - For some reason only '70s RCA records developed mold on the labels.
- - - a few polystyrene records were played with a stylus force too high for styrene.

- Most of the tapes I owned in the Quad era have failed.
- - - Many of them had the oxide fall off the tape.
- - - Others had the adhesive holding the oxide glue the layers of tape together.
- - - The rubberized nonslip backings deteriorated and stuck to the tape in places.
- - - Many of the beginnings of reel tapes broke at the end of rewind and snarled.
- - - The splices in some 8-tracks and cassette leaders failed.

- The VHS tapes failed in similar ways.

- Many of the floppy discs I had in the 1980s-90s have failed in the same ways.

- I have had very few CD and DVD failures.
- - - A very few developed air leaks that corroded the foil sheets.

- I have had a small portion of my CD-R, DVD+R, and DVD-R recordings fail:
- - - Some early CD players could not play certain brands of CD-R.
- - - Several early CD-Rs were destroyed in hot cars.
- - - Some newer DVD players won't play the DVD-R discs.

- Many of the audio computer files I made in the 1990s were lost because Microsoft deprecated the formats and didn't tell anyone.
Tu summarize you say: nothing last forever.
 

quadaholic

Active Member
Joined
Feb 2, 2010
Messages
78
Location
Newtonville MA
The base matrix for UD4 was the BMX matrix (often called UMX).

QMX was UMX with 4 separate channels of transmission and complementary versions of the BMX 2-channel version for the other 2 channels. QMX would reproduce a discrete reproduction of the 4 original channels when properly decoded.

UD4 was a version of QMX with the carrier channels restricted to 3KHz bandwidth.

This is all covered in the article:

"Discrete-Matrix Multichannel Stereo" D.H. Cooper & T. Shiga,
Journal of the Audio Engineering Society 06/1972 V 20 pp. 346-360

It does require a bit of mental translation since Denon changed the product names in the interim.

These three articles show the encoding of all matrix systems:

"Analyzing Phase-Amplitude Matrices" Peter Scheiber,
Journal of the Audio Engineering Society 11/1971 V 19 pp. 835-839

"4-2-4 Matrix Systems: Standards, Practice, and Interchangeability " John Eargle,
Journal of the Audio Engineering Society 12/1971 V 20 pp. 809-835

"A Geometric Model for Two Channel Four Speaker Matrix Stereo Systems" Michael Gerzon, Journal of the Audio Engineering Society 03/1975 V 23 pp. 98-106

Notice that QS, EV, DQ, and Scheiber are all minor variations on the same matrix encodings (with the general characteristic called RM).

Notice that RM, BMX, H, and Ambisonics are four different matrix systems.

As long as CD-4 requires the equivalent of a clean-room environment to work properly, I do not consider it a viable system for general public use.

Note that later CD-4 records were made with a variable level carrier. The carrier level decreased when the audio level increased so the record was not overcut. This could cause Q-eight's problem.

Radio stations hated CD-4 because it could not be slip-cued - the blank groove is not blank between tracks.

Tape as the only format? Not for longevity:

- I have very few records (out of 1000s) that have failed (the oldest is from 1908).
- - - I have a couple of World-War-II-era records that delaminated and fell apart because substitute materials were used to make records due to rationing (the cores were corrugated cardboard).
- - - All of my CD-4 records were ruined when I got them (I was a college student at the time and could buy only used records - I was given the cartridge and demodulator).
- - - For some reason only '70s RCA records developed mold on the labels.
- - - a few polystyrene records were played with a stylus force too high for styrene.

- Most of the tapes I owned in the Quad era have failed.
- - - Many of them had the oxide fall off the tape.
- - - Others had the adhesive holding the oxide glue the layers of tape together.
- - - The rubberized nonslip backings deteriorated and stuck to the tape in places.
- - - Many of the beginnings of reel tapes broke at the end of rewind and snarled.
- - - The splices in some 8-tracks and cassette leaders failed.

- The VHS tapes failed in similar ways.

- Many of the floppy discs I had in the 1980s-90s have failed in the same ways.

- I have had very few CD and DVD failures.
- - - A very few developed air leaks that corroded the foil sheets.

- I have had a small portion of my CD-R, DVD+R, and DVD-R recordings fail:
- - - Some early CD players could not play certain brands of CD-R.
- - - Several early CD-Rs were destroyed in hot cars.
- - - Some newer DVD players won't play the DVD-R discs.

- Many of the audio computer files I made in the 1990s were lost because Microsoft deprecated the formats and didn't tell anyone.
I was never aware of corrugated cardboard being used for records during wartime. Wow!
 

Wurly1

Active Member
Joined
Dec 15, 2019
Messages
56
Location
Canada
Matrix CD-4 would work but i dont see how it could be Stereo-compatible.
As for CD-4, i have a reliable decoder system (and my room is not clear lol) but i wish they had made them Non-Stereo-compatible. The standard groove signal is Front + Back while the modulated signal is Front - Back, because the final matrix of those signals introduce noise on both when either signal is damaged. I would have preferred the standard groove to be Front and the modulated signal being Rear. On the other hand worn-out CD-4 stereo-comatible records will continue to play perfectly on a standard system.
UD-4 seemed to be better but not Stereo-Compatible and came too late.
Final note: they had to make choices when they created it and personally i think it's a miracle this analogue discrete quad system work at all !
 

MidiMagic

400 Club - QQ All-Star
Joined
Jul 5, 2010
Messages
422
Oh, yes. Shellac was used to make 78 rpm records.

Japan captured most of the sources of shellac, a resin secreted by the female lac bug on trees. Since shellac is an insulator for electric wiring needed for warplanes and RADAR, shellac was rationed from 1939 to 1946 in most western countries.

This caused the following shellac-conserving limitations to appear in these countries:

- In order to buy a new record, you had to turn in a used one. It didn't matter if the used record had broken. The shellac was melted down to make new records.

- Easily worn records of Vitrolac (vinyl) were sold at reduced prices. They lasted only a few plays.

- Most record companies used good shellac for the recorded surfaces of the records and sandwiched a filler material between the recorded surfaces.

- The following materials were used for the filer materials: shellac mixed with paper or sawdust, sawdust mixed with hide glue (standard wood glue), asphalt, Bakelite, cardboard, paperboard, Masonite, Fiberglas, plastic, and even glass.

- Some record companies mixed shellac with hide glue and paper dust to extend their supply of shellac.

- Because aluminum was also rationed, lacquer masters were made with a glass substrate.

These records made of substitute materials broke more easily. The record changers of the 1930s broke many of the new records.. This was the main reason record changers stopped using knife-type record dropping and started using push-type record dropping. A few changers were made that lowered the records instead of dropping them. And the push-type changers had an s-shaped spindle to slow the record as it falls.

Most of these substitute records have become even more brittle with age. They should not be dropped from record changer spindles.

In 1947, with the development of the LP, vinyl started replacing shellac as cartridges appeared that did not damage vinyl.
 

MidiMagic

400 Club - QQ All-Star
Joined
Jul 5, 2010
Messages
422
The matrix encoded CD-4 was a combination of matrix and CD-4 with the disadvantages of both.

I wish you would stop repeating this false assertion. My listening room is anything BUT a clean room and I listen to CD-4, successfully, all the time.

Doug
In my experience, it is not false. Air pollutants beyond my control were a problem where I lived as a teenager and college student:

- I was required to share (let my brothers play my records). They were anything but careful with them.

- My father was into woodworking. Often the air was filled with fine sawdust.

- One of my brothers was an actor. Often the air was filled with makeup powder.

- I had no budget for new records, I bought used records. Every CD-4 record I bought was already ruined when I got it. Even the one that came with the demodulator was already ruined when I got it.

- Standard record cleaning products removed the dust from stereo records, but I was never able to get any of those CD-4 records to work. Snap-crackle-pop.

I still have three of those records and a working AT-12 cartridge.

How do YOU keep this stuff from happening? Is there a way to get all of that crud out of the grooves?
 

Soundfield

900 Club - QQ All-Star
Joined
Dec 29, 2011
Messages
935
Location
Essex, UK
SQ was notoriously bad when it came to stereo and mono compatibility; QS, while not perfect, was better for that.
If you look at the matrix encoding coefficients you will see that SQ was actually better for stereo compatibility from a front channel point of view (as CBS wanted). Actually, it was one of SQ's few advantages! But it came at the consequential cost of front - rear and L/R rear separation. Granted, mono compatibility for the rear channels was indeed poor. QS on the other hand compromised front channel separation in order to create a more uniform channel separation overall (hence 'Regular Matrix'). Interestingly, the BBC found the stereo and mono compatibility of both SQ and QS to be so poor from a broadcasters perspective that they rapidly abandoned both formats early in their laboratory trials and started on the path to develop Matrix H / HJ. Even that failed to fully solve the problem.
 
Last edited:

Marcsten

400 Club - QQ All-Star
Since 2002/2003
Joined
Sep 24, 2002
Messages
403
Location
Seattle
If you look at the matrix encoding coefficients you will see that SQ was actually better for stereo compatibility from a front channel point of view (as CBS wanted). Actually, it was one of SQ's few advantages! But it came at the consequential cost of front - rear and L/R rear separation. Granted, mono compatibility for the rear channels was indeed poor. QS on the other hand compromised front channel separation in order to create a more uniform channel separation overall (hence 'Regular Matrix'). Interestingly, the BBC found the stereo and mono compatibility of both SQ and QS to be so poor from a broadcasters perspective that they rapidly abandoned both formats early in their laboratory trials and started on the path to develop Matrix H / HJ. Even that failed to fully solve the problem.
Yeah, in the days of crappy decoders, the deal was SQ was better left/right, and QS was better front/back. And CBS in particular made it a point that stereo compatibility was almost more important than how it worked in quad. So the mix had to be consistent. If something was in the left in the stereo release, in SQ it was in the left front, period. Other labels didn't hold to this.
 

jaybird100

1K Club - QQ Shooting Star
QQ Supporter
Joined
Apr 27, 2006
Messages
1,263
Location
Pembroke Pines, FL
If you look at the matrix encoding coefficients you will see that SQ was actually better for stereo compatibility from a front channel point of view (as CBS wanted). Actually, it was one of SQ's few advantages! But it came at the consequential cost of front - rear and L/R rear separation. Granted, mono compatibility for the rear channels was indeed poor. QS on the other hand compromised front channel separation in order to create a more uniform channel separation overall (hence 'Regular Matrix'). Interestingly, the BBC found the stereo and mono compatibility of both SQ and QS to be so poor from a broadcasters perspective that they rapidly abandoned both formats early in their laboratory trials and started on the path to develop Matrix H / HJ. Even that failed to fully solve the problem.
Obviously, any matrix system was going to be a compromise. SQ may have maintained left-right front separation, but the rear information tended to fold in between the two stereo speakers. This created more possibility of signal cancellation when listened to on a mono receiver. QS did narrow the front image somewhat, but the rears tended to project out beyond the stereo speakers, creating a wider soundfield. In mono, center back signals would still cancel, but there was less loss of rear channel information when listened to monaurally.

Another point, too, is that if QS were so bad for compatibility, why then did so many more FM stations in the US opt to go with QS encoders than SQ? I worked at a station that had initially purchased an SQ encoder from CBS Labs, only to pull it out and replace it with a Sansui QSE-5B. I did some of the encoding from quad tapes, and the results with the QS encoder were noticeably better than with the SQ. Comparisons were made listening in quad, stereo, and mono. The QS was better than the SQ. Our format was adult contemporary; even running stereo material through the encoder's stereo enhance mode sounded great. That was our experience.

I remember seeing an ad in a British audio magazine for an adapter that would allow the BBC's Matrix H broadcasts to be received on systems with QS or RM decoding. How well it worked, I don't know. I have an LP, released by Virgin Records, that's in the Matrix H system. I'll only say that it's "interesting".
 

Soundfield

900 Club - QQ All-Star
Joined
Dec 29, 2011
Messages
935
Location
Essex, UK
The BBC really disliked SQ and weren't especially keen on QS. There were many reasons for this, but foremost amongst them were both formats' variously poor mono and stereo compatibility, lack of accurate image localisation and audible phase related artefacts. I'm sure you'll have read the BBC's detailed study reports into the performance of the currently available matrix systems. I don't think any other broadcaster did as much research (if any) into the subject as the BBC and I've no idea why some US stations made the decisions that they did. What the BBC really did like however, was Sansui's encoding and decoding technology, particularly Vario Matrix. They used Sansui equipment as the platform on which to to host the development of something based rather on more on QS than anything else - Matrix H.This had its own peculiarities, and quickly morphed into Matrix HJ and its subtle variants. Commercial Sansui equipment was therefore capable of being simply modified to decode Matrix H and HJ. Sansui released a couple of such decoders for sale in Europe but it was all rather too late as by then the BBC's year long on-air demo of HJ was coming to an end.
 
Top