SQ decoding?

[rcf]

Any possibility of adding the capability to
decode SQ or QS quadraphonic signals? This could be a very fun and useful option.

[Craig Maier]

Re: SQ decoding?

It has been a long time since I have heard of these encoding schemes, but if I remember them correctly, they use a ultrasonic subcarrrier for the two additional tracks. Those subcarriers are around 25 to 35 KHz I think. That would mean that most sound cards would not be able to deal with them. One would need a 96 KHz sampling system which is not presently real common.

[rcf]

Re: SQ decoding?

<BLOCKQUOTE><font size="1" face="Verdana, Arial">quote:</font><HR>Originally posted by Craig Maier:
It has been a long time since I have heard of these encoding schemes, but if I remember them correctly, they use a ultrasonic subcarrrier for the two additional tracks. Those subcarriers are around 25 to 35 KHz I think. That would mean that most sound cards would not be able to deal with them. One would need a 96 KHz sampling system which is not presently real common.<HR></BLOCKQUOTE>

There were at least 3 different standards for quadraphonic encoding on LPs. The one you describe is called 'CD-4' and works much as you say. This is a discrete system; the back two channels are transferred to a higher frequency range and special cartridges and styluses are needed to play them back into a demodulator.

The more common quad systems encoded the back two channels into the regular stereo channels so no special equipment beyond a decoder is necessary to play them back. The SQ system used one set of transformation matrices that emphsized maintaining regular stereo separation and SQ LPs can be played on regular stereos. The QS encoding has less left/right separation & plays poorly on standard stereos.

What would make this interesting nowadays is that (1) old quad gear has become exhorbidantly expensive; (2) with new home stereo equipment, 4(+) channel audio has become fairly common; (3) DVD players that are DTS compatible can play CD-Rs with DTS 4 channel audio made from quad LPs; (4) there is existing commercial software available to make DTS CD-Rs; (5) thus, old quad LP collections can be preserved and enjoyed beyond the life time of the electronics, which are at least 25 years old nowadays.


The encoding formulas are fairly simple.
SQ:

C = sqrt(2)/2

L'= LF + C*(LR[-90] + RR)
R'= RF + C*(RR[90] + LR[180])

LF' = LF + C*(LR[-90] + RR)
RF' = RF + C*(RR[90] + LR[180])
LR' = LR + C*(LF[90] + RF[180])
RR' = RR + C*(RF[-90] + LF)

in which the [theta] is a phase shift in the signal by theta degrees.

QS:
L'=0.924(LF+LR[90])+0.382(RF+RR[90])
R'=0,924(RF+RR[-90])+0,382(LF+LR[-90])

LF' = LF + C*(RF + LR[90])
RF' = RF + C*(LF + RR[-90])
LR' = LR + C*(RR + LF[-90])
RR' = RR + C*(LR + RF[90])

[Craig Maier]

Re: SQ decoding?

Thanks for the very informative tutorial on this subject. I was not aware of the additional permutations of this technology as you have described it. Interesting. I wonder how much of this stuff is out there? Anyway, we will study the equations you have graciously presented to us and think about the commercial aspects of your proposition. It certainly is intrigueing to me.

Regards,

Craig Maier

[Craig Maier]

Re: SQ decoding?

BTW - having just looked at the equations, performing the "all pass" filtering should not be a particular problem. The only issue is the commercial viability of this proposal. Convince us that there is enough stuff out there to decode this.\\

Craig