Archive | 11:13

A Matter Of Phase

7 Jul

As a sort of preface to my small overview of available methods of decoding the two most commonest PA quadraphonic matrix systems used in the 1970’s on LP i thought i’d prime those interested by trying to put a few thoughts on the subject of what is a PA Matrix, and what does a decoder do.

The addition of 90 degree phase shifting to the limited Amplitude Matrices DY & EV-4 opened up the possibilitiy of finally being able to encode (which was never a problem) and decode (alwaus a problem) a four channel recording with a fair resemblance to the original, even though the hardware wasn’t really up to the job, the possibilty was there.

So there’s much to be thankful off, dear old phase proved to be a very useful tool. So why then do those who think they know about the PA Matrices choose to completely ignore it in their attempts to create a process to decode them?

Phase Amplitude, sort of says it all really. A seemingly simple idea that seems to be confusing enough for people to apparently ignore how the matrix employed, in which ever system is being focused on, actually works.

Again, i must here stress that when i use the word ‘decoder’, i mean the device that mathematically decodes the mathematically encoded stereo music track BEFORE it is fed (if fitted) to whatever form of seperation enhancement is used.

So, what are the functions of a PA matrix decoder?

Firstly, and you’d think this was blatelently obvious, it is to accurately return the encoded stereo source to match, both in amplitude and phase, the original four channel source.

‘Simples’ (copyright ‘Alexander’ meerkat) you might think.

The reason for this process to be as accurate as possible is to ensure that sounds are repositioned in their correct locations, both in amplitude and phase, irrespective of the large amount of unwanted crosstalk information that is left over from the process. It’s a well know fact that, basically, the maximum amount of front/rear seperation possible is around 3db. Not a lot really, and depending on the mix it’s possible to gain a fair impression of the mix.

This of course means there’s a lot of unwanted rear information on the fronts, and vice-versa. But all’s not lost as one of the positives here is acoustic cancellation.

Because of the nature of how the decoding process works, by returning the rears back to their original positions the unwanted information is also phase adjusted, and dependant on various issues like placement within the mix, etc, there can be an acoustic enhancement of the seperation that doesn’t show up on electronic measurements.

Once the four channels have been decoded they are then send off (if fitted) to the seperation enhancement section. It is here that the accurate decoding of the source is eessential. If correctly designed it will be working on the theoretical positioning information as per that system.

Obviously any errors made in the decoder section are magnified here.

So, that’s the first function (hopefully) taken care of, but what of the second function? Almost as important as the first it’s……..

And it’s here where i leave it to those who know better than me to tell you all what the second function is……….. 😏

Hopefully this will have primed you for some upcoming reviews of software attempts at decoding our beloved PA matrices.

%d bloggers like this: