Hi all - again Firstly, i must appologise for continuing on this line, but i'll admit my maths isn't up to the level of you guys, but i've found something that, to put it mildly, has gotten to me.
And that is the Matrix 'H' maths. This is my re-write of what i believed to be the equation to use when working on doing a software decode of 'H' taken from 'The Bible': Quadraphony:developments in Matrix H decoding (BBC RD 1977/2), which is said to be a "complex conjugate of the encode matrix". Lt = 0.940 j-10, 0.342 j-65, 0.940 j25, 0.342 j115 Rt = 0.342 j65, 0.940 j10, 0.342 j-115, 0.940 j-25 Now, reading a paper issued by the BBC called: Quadraphony - Matrix H Data, and which states it is based on information the previously stated paper, it has what is stated as "Matrix H transfer characteristic" (written as 2 channel to 4) being as: Lt = 0.940 j10, 0.342 j65, 0.940 j-25, 0.342 j-115 Rt = 0.342 j-65, 0.940 j-10, 0.342 j115, 0.940 j25 And i'd to throw in the recently aquired 'HJ' equation (thank you *******) which is written in the same way as above - 2 to 4, and the QS encode equation - you'll see why: Lt = 0.966 j2.8, 0.259 j67.3, 0.940 j-27, 0.342 j-103 Rt = 0.342 j-67.3, 0.940 j-2.8, 0.342 j103, 0.940 j27 Lt = (0.924 Lf + 0.383 Rf) + (0.924 Lb j + 0.383 Rb j) Rt = (0.383 Lf + 0.924 Rf) + (0.383 Lb k + 0.924 Rb k) j = +90 degrees k = -90 degrees I'd now like to add this piece of information into the boiling pot: The BBC released (at the time) a modification to QS decoders to enbale people to 'enjoy' the encoded broadcasts with consisted of applying a 60 (58?) degree shift to the right channel, and then both channels were swapped over before being fed into the decoder, and swapped back after decoding. Then there's the reason why i'm asking: the process i use to decode phase matrix encoded material relies on the removal of the unwanted phase information, thereby leaving the wanted, which works remarkably well on SQ and QS. So you can see i need accurate information. Right, now down to the nitty-gritty.... As can be seen from the 'H'equatons i've listed, there's a disagreement in the polarity of the phase. When you then throw in the way that the QS decoder has to be modified it gets a little confusing, Using the equation on the 1979/2 document as the 'encode' matrix, Adding +60 would appear wrong if the equation is to be believed (and the +60 really is that, and not -60) as it doesn't bring the Rb near the required +90. BUT doing -60 would bring it close to the required -90, but going along that road then brings into question why the swapping of the channels? or am i missing something here? The other question i have to ask is whether the 2nd 'H' formulae is to be ignored, and the similiat;y written 'HJ' formulae is to be re-written with the phase polaritys swapped? I now realise i need to go back to learn maths, but i have to say this is driving me a little nuts Richard -------------- next part -------------- An HTML attachment was scrubbed... URL: <https://mail.music.vt.edu/mailman/private/sursound/attachments/20120130/bdec2181/attachment.html> _______________________________________________ Sursound mailing list [email protected] https://mail.music.vt.edu/mailman/listinfo/sursound
