Archive | 13:09

More What’s,Whys And Wherefores of Decoding, And Three QS “Test” Releases

9 May

Since closing both blogs last year, I’ve spent some time doing some improvement work on all of the matrix decode process’s, one ending up quite a major upgrade and the others a small, but essential, alteration.

At around the time of the blogs closure, Owen made an off-the-cuff remark that it was a shame i hadn’t spent as much time on QS as I had done on SQ. He was correct, of course, my personal dislike of QS, and my preference for SQ, showed in my work and the number of releases (especially in the early days). So began my final work on the decoding process’s for the DY/EV-4, QS & SQ matrices.

The upgrade to the QS process (QS2d) is quite marked, ending up nothing like earlier, or other, versions. I “believe” that finally a QS decode can now hold it’s head up against that of an SQ (SQd2) decode in terms of separation, image stability and low level detail.

Those eagle-eyed amongst you may have noticed the upgrade to the SQ decoding process. The upgrade is solely devoted to solving a previously unknown error in the decoding process in decoders based on the official CBS/Sony specs.

I will admit that there is the possibility that many may not notice much of a difference between an SQd1 and SQd2 decode of the same album, it’s partly down to whether you listen, or you . But the improvement is there, none the less, solidifying the overall image like never before.

As mentioned, this upgrade corrects an error in the original decoding mathematics as developed by Ben Bauer, and is inherent in all “official” decoders (be it discreet or based on the Motorola chip set) that introduces a phase error of around 25 degrees on the rear right during the decode process.

If the mathematics of the whole decoding process are checked, all appears correct, but as any engineer who has been in the electronics field for long enough will attest, there are times that unpredictable things happen when it comes to the practicalities of a working model.

My work, and conclusions, were checked not only by using my decoding processes but by the decoded output from a Tate decoder, which showed exactly the same error.

The answer to the question “Why didn’t they see the issue at the time” is simply that they just didn’t have the tools we have today. Many of the tools we take for granted now could probably never even be imagined of back then.

At last SQ is capable of delivering the kind of performance that Ben Baur knew it was capable of, delivering a phase accurate representation of the original master.

Having been able to overcome this long standing issue, I believe that the best, and only worth-while (to date), SQ decoder (The Tate) can be improved by modifying it so it can comply with the above (new) standard, which would not only deliver a better decode, but the logic section should be able to operate more efficiently, or as it was meant too, producing better image stability with less pumping artefacts, etc.

DY/EV-4 decoding has also¬†been tweaked, although it’s impossible to make major in-roads when it come to decoding these two pre-phase matrices, simply because they are just too darned basic.

As a demonstration of QSd2 I’m making available the following three albums, which luckily have two different types of mixes. These are the “Four To The Wall” type that gives an artificial presentation much loved by the majority of quadraholics, and the “Live In The Studio” mix, more often than not, used by jazz labels which attempts to create how things were in the studio when the album was recorded live.

QSd2

Due to an issue with DTS and Dolby Digital audio, this is now reissued

Advertisements
%d bloggers like this: