"new" way to decode SQ

[Disclord]

so with LF from LF , RF from RF, LB from LS,RB from RS....and if you then disconnect the rest of the speakers and perhaps move the LS and RS back a bit would that be acceptable?...

and does the CF really need to be on since it's not part of the sq quad system...

it also makes me wonder if some brands work better than others

cheers

CF is an actual "encoded" position, even if no decoder provided a 'hard' speaker output for it - I personally prefer the sound of an actual "center" VS a phantom center.

There DO seem to be differences in matrix decoding between brands/with time - I was recently given a Denon 3808 receiver to replace my Kenwood Sovereign THX receiver that totally died, and the Denon's Pro-Logic The Sequel decoding is very 'different' sounding that that of the Kenwood. The Denon is "tighter" in some ways, but I can hear dialog pull to the rear and snap back more more than the Kenwood did - it's the same with DTS Neo:6 - but in that case, the 'singing along' artifact of DTS Neo, due to the sub-band decoding, is totally gone, thank goodness - I LOVE Neo, but hated the weird birdies and tones it created. From reading other forums and such, it appears that Neural's decoding has been tweaked a bit over time. Of course, it's not just confined to Dolby, DTS and Neural - Lexicon has done HUGE amounts of tweaking with Logic-7 - enough tweaking that the decoder should really be renamed because it bears almost no resemblance to the original Logic-7.

 

[Disclord]

Couldn't wait to give this a try on my Denon. I think the mechanics are there, so the quality of the decode may have more to do with the source material. I tried Columbia's cast albums for My Fair Lady (20th anniversary production) and ThreePenny Opera (with Raul Julia) and got fantastic separation, very discrete. Next was Neville Marriner doing Eine Kleine Nachtmusik (Angel S-37443) and got some ambience but it was hard to tell if it was intentional or just reverb. Finally was an Italian pressing of Mozart's Symphony No 29 with Leopold Hager conducting (PDU ACSQ 60071) which had nothing. It was almost mono. Would love to hear a report from someone who can try a Vanguard or Project 3 release.

Which Denon do you have? I have the most up-to-date firmware in my AVR-3808, but I haven't updated to the Audyssey Volume & Audyssey Dynamic EQ that Denon has made available - neither interests me, although I absolutely LOVE the Audyssey system and what it's done for my speakers. I have a Project 3 SQ LP of Tony Mottola that I'll give a whirl thru Neural - it's a super Quaddy mix that sounds fantastic via the Fosgate Tate II, so I'll see what Neural does with it. I've noticed that Neural does have a tendency to make things sound quite mono sometimes... I don't know why. One SQ LP I need to try is Barbra Streisand In Concert - it has, at least in the 3 pressings I've owned, basically NO surrounds encoded at all - it's all front stereo, and one of the most pathetic/worse quad releases ever done - imagine if that was someones first SQ LP heard? They'd think the system didn't work at all.

 

[boojidad]

Mine is the AVR-1708. It has the Audyssey processing but I'm not sure how much of a difference it makes. I'm thinking of raiding some local thrift shops to see how much SQ I can get my hands on now that I have something that actually works. Somebody stop me! )

 

[oxforddickie]

"One SQ LP I need to try is Barbra Streisand In Concert - it has, at least in the 3 pressings I've owned, basically NO surrounds encoded at all - it's all front stereo, and one of the most pathetic/worse quad releases ever done - imagine if that was someones first SQ LP heard? "

Interesting, i had that down for a release in the early days, but thought there was something wrong. Think i'll have another look at it

 

[surroundophile]

so with LF from LF , RF from RF, LB from LS,RB from RS....and if you then disconnect the rest of the speakers and perhaps move the LS and RS back a bit would that be acceptable?...

and does the CF really need to be on since it's not part of the sq quad system...

it also makes me wonder if some brands work better than others

cheers

As answered above, CF is part of the SQ system. That said, you can turn it off and have a phantom center.

As far as LS and RS, Left Side and Right Side are phantom (virtual) channels. They exist because the intended signal appears in mono in the LF & LB as well as the RF & RB channels. So if you disconnected the LB and RB channels, you would only have 2 or 3 channels (depending on if you have a CF). Besides, who would want to do all the modifying just for SQ LP's). With my prepro, the back channels didn't decode at all!

 

[oxforddickie]

Sorry, but CF IS not and never was part of the SQ or QS matrix system. They are Quadraphonic systems, i.e; 4 channel. Because of the complex phase issues involved, decoding any other way will produce incorrect results.

 

[Disclord]

Sorry, but CF IS not and never was part of the SQ or QS matrix system. They are Quadraphonic systems, i.e; 4 channel. Because of the complex phase issues involved, decoding any other way will produce incorrect results.

Yes, Center Front was a legitimate encoding point for SQ - the SQ position encoder had it, and it's mentioned in many of CBS's (and others) AES papers. Now, it may not have had a decoded hard-speaker position in any produced decoders, but considering it was a legitimate encoded signal, anyone could have made a decoder that gave it a 'hard center'. Both Ben Bauer and Greg Badger talk about speaker layouts and decoding to derive just such a 'hard' center. The SQ system was NOT limited to only 4 decoded speakers - the SQ system had a precise phase/amplitude relationship for EVERY decoded direction and decoders could be derived that gave a speaker to those locations. If it is not a part of the SQ matrix, why did all full-logic decoders and the Tate DES have a "Center Front" control signal to signify a signal there and its removal from the rears?

 

[oxforddickie]

Whether or not the possibility existed in the encoders, i cannot find it believable any recordings were so encoded. They couldn't make a decoder capable of anything like a reasonable replication of the original four channel tape, let alone trying to push it to five speaker, especially as there were no five channeldecoders/amps/receivers available to the consumer.

Also, i think you've answered your own question regarding the "center Front" control....... to remove it from the rears due to the inability of doing it any other way.

It would actually be good to see some proper descusion here about the real issues of decoding, rather than irrelevant rumour mongering that does no real purpose in the improvement in decoding.

But then, who am i?

 

[Disclord]

Whether or not the possibility existed in the encoders, i cannot find it believable any recordings were so encoded. They couldn't make a decoder capable of anything like a reasonable replication of the original four channel tape, let alone trying to push it to five speaker, especially as there were no five channeldecoders/amps/receivers available to the consumer.

Also, i think you've answered your own question regarding the "center Front" control....... to remove it from the rears due to the inability of doing it any other way.

It would actually be good to see some proper descusion here about the real issues of decoding, rather than irrelevant rumour mongering that does no real purpose in the improvement in decoding.

But then, who am i?

EVERY SQ recording with a vocalist placed at equal level between the front speakers only and in phase was encoded with "Center Front." Have you never read any of the AES papers or related patents? It's encoded the same way in Dolby Surround and QS. Look at the Scheiber Sphere for SQ and compare it to QS and Dolby Surround on the Sphere.

Center Front was recorded just like stereo (and stereo certainly has a 'center front') - equal levels in phase to both Lf and Rf channels with no phase-shift (relative phase shift in SQ is not related). To remove it from the rears in decoding, the CBS's F/B Vari-Blend Logic (and Tate DES and Shadow Vector worked this way too, but went about it in different manners) did so by summing the two rear outputs, since CF is in phase opposition to the Center Back, which, by rotating Lb and Rb to CB (as seen on on the Scheiber Sphere), the Lb/Rb become equal, thus cancelling the orthogonal original.

The "Center Front" Control is the directional detectors signal indicating that a predominant signal exists there, it says nothing about how its crosstalk is removed - its up to the rest of the decoder to apply the correct coefficients to ensure that CF's leakage components (which, in SQ are into the rear channels) are removed from them.

I'm not "rumour mongering" not by ANY stretch of the imagination - I'm discussing something that is well known about SQ - for you to say SQ doesn't have (and no one used) an encoded Center Front on SQ recordings is jaw dropping. You DO know how SQ works, don't you?

For anyone who doesn't believe Center Front exists on an SQ recording, simply play it through a Dolby Pro-Logic or DTS Neo:6 decoder - you'll here it there, from the Center Front speaker, in all its encoded glory!

 

[oxforddickie]

Stramgely enough, i have. What i said was basically all this talk does nothing to improve the correct decoding of the Quadraphonic systems AS THEY WERE MEANT TO BE HEARD.

Oh, you obviously don't know me, i DO know how SQ works, AND how to decode it, AND to overcome some of the issues.

i shall continue looking for a descent conversation..................


OD

 

[Disclord]

Stramgely enough, i have. What i said was basically all this talk does nothing to improve the correct decoding of the Quadraphonic systems AS THEY WERE MEANT TO BE HEARD.

Oh, you obviously don't know me, i DO know how SQ works, AND how to decode it, AND to overcome some of the issues.

i shall continue looking for a descent conversation..................


OD

All we were talking about in this thread, before you came in with the blanket statement that Center Front didn't and never did exist in SQ encoding, which is wrong, no matter how you want to defend it, was getting kind of interesting results with Neural - not totally accurate by any means, but better than any other parametric-type decoding except for the Tate System.

Now, I'm not asking this question flippantly or anything, but you say you know how SQ works - why didn't you seem to know that equal level, in-phase signals to the front L/R are classified as "Center Front" (and always were) and used in the majority of SQ recordings with lead vocals, and can be decoded as such? Contrary to what was claimed by some (Michael Gerzon), SQ encodes a full 360 sphere of sound and is NOT limited to decoders that produce only the 4-speaker layout of 'quad' - a system like Shadow Vector or the Tate DES could be adapted to create correctly decoded speaker feeds for Center Front, Center Back, Left Center Side and Right Center Side, in addition to the usual Lf, Rf, Rb, Lb of quad. A "Center Top" speaker feed could also be created. Sadly, at the time of those decoders creations (Tate, etc), additional logic-derived speaker feeds were not the norm, so in terms of dedicated SQ decoders, only the typical "quad" speaker layout was ever produced.

This is kind of unrelated, being a PC decoding question... I'm not wrong in thinking you've done PC-based SQ decoding, right? If I am wrong, ignore the rest.
Do you have the SQT-1100 CBS test record for SQ? With its test tones, both the static and logic dynamics ones, when the Center Front one is on, and PC decoded, how does the decoded result sound? Does it image in a 'sharp' phantom or is is blurred between the front speakers? I ask because the PC conversions I've downloaded (like Paul Simon's) all have absolutely TERRIBLE phantom imaging - even the sides and back have no in-phase phantoms like they should. The four corners are reproduced OK, but thats it. So I've been trying to find out if that's a problem with PC based decoding or if its an anomaly of those specific recordings PC processing. The Fosgate Tate II produces absolutely rock-solid phantom imaging that's not blurred or 'widened' in any way. I haven't been able to find a PC decoded SQ recording that duplicates that.

 

[oxforddickie]

No, i don't have the test record, such things are rarer than dodo's here in the UK.

You obviously arn't aware of me, and i'm not going to brag but yes, but i've done a little SQ / QS / UHJ / Matrix H (still in alpha) decoding on the pc in my time (lol), and i believe i've pushed the limits of PC decoding to it's limits (well as of today anyway - there's always something that just needs tweeking....). I still have "tentative plans" to go six channel with my series of scripts (sorry, not avaialable to non-releasers by the way), but that's going to be after i put all of the other matrix systems in order. But there is no way it will include a center channel.

Of course i know about the "center channel", learnt that in college in 1970. It's always been there, even in stereo, BUT it is not (and this was my point) a recognised channel in the SQ or QS maxtrix's. Just because you "could" decode it, doesn't mean you "can" successfully.

Attempts to produce a proper center channel is nothing but a mine field, and to be honest, will do little to improve reproduction. better off with the sides, do you not think?

I'm sorry i interupted your talk about " "a new" way of decoding SQ. To be honest, i don't think it'll be that accurate - shame though, would love to have a good excuse to stop working on improving the decoding of the matrix's........ and actually listening to it instead........

 

[inquadwetrust]

OD - I'm familiar with your work through your blog, I've no rep here (or elsewhere hopefully) just someone who loves surround sound. Your decodes from QS etc material show your thorough understanding of what is required. This coupled with your pursuit of feedback and your commitment to re-encodes when scripts have improved is remarkable.

Disclord - I yield to both yourself and OD's superior skills and knowledge of all matters surround but I have listened to some real hum-dingers (both official and un-official) and I feel OD's work is genuinely worth a listen and if you haven't already have a look see at the URL in his post footer.

In these lean-picking times surround junkies like myself rely on the hard work put in by people like OD, Romanotrax and so on. I've no doubt that some of the releases may not pass muster with those who own the original equipment and recordings but some of those releases get more play than commercial DVDAs I own.

In hindsight I don't think DTS understood my enquiry at NAB - I stressed quad matrix recordings were a side-line interest (professionally I deliver a lot of 2D/3D DCP content for digital cinema and the neural upmix is something clients are interested in) and the demonstration material was a short car chase section from Die Hard 4 and then a music clip from The Killers Live. The Killers audio was a straight stereo PCM recording from a DVD that was then processed in real-time from Pro-Tools into 5.1. Emphasis could be changed with regard to the sub and Front and Back presence and it sounded good. Discrete with no pumping or whooshiness going on. But as a seasoned demonstrator of software/hardware for years, I know I'd pick something which would work and work well.

Gentlemen - I salute you both, your knowledge and experience is gratefully received.

 

[oxforddickie]

Thanks for your comments IQWT, as your aware quality has always been top of the list, nice to see the hard work appreciated. As a side line i'm sure i'm not giving too much away but it's hoped to include more FM broadcast's in the very near future, especially as "PPR" allows improvements in what can be pretty dodgy recordings to be decoded more successfully (sticking my kneck out there a bit LOL)

OD

 

[Disclord]

In this thread there was a mention that the Neural decoding in one receiver might be different than the Neural decoder in an earlier model - I found a post on AVS Forum from Roger Dressler (Dolby Labs) about that very subject and thought I should repost it here. I can't find the link to his comments - I copied them to a Word document but didn't copy the URL... yes, I'm an idiot! Anyway, here is what he wrote about it:

Roger Dressler discussing Neural & DTS Neo 6

They may not have publicly said so, but Neural has changed quite a bit, at least based on my extended comparisons made in 2007 (in a Yamaha AVR) vs 2010 in the Sherwood 972. It's not unusual, Neo:6 went through at least 4 different code revs over its life. Improvement is fine, but it poses problems for people comparing their experiences--as they may well be listening to different versions.

 

[ghalteman]

Here's a big question... I've been using the old SQ decoding software with Adobe Audition 1.5. Can anyone provide a link to the thread that shows what is needed to run the latest and best SQ decoding method? The whole process? Thanks.

 

[Old Quad Guy]

Here's a big question... I've been using the old SQ decoding software with Adobe Audition 1.5. Can anyone provide a link to the thread that shows what is needed to run the latest and best SQ decoding method? The whole process? Thanks.

The latest release is the "SQ Plus" provided by oxforddickie on 9-21-2010:

SQ Plus - Improved script available

https://www.quadraphonicquad.com/forums/showthread.php?13482-SQ-Plus-Improved-script-available

But it works only with AA 3.0 AFAIK. I'm not sure if the script could be modified for AA 1.5. But we could ask. You'll also want to get the free VST plugin "Phasebug" at bottom of this link: http://www.betabugsaudio.com/plugs.php

 

[DKA]

One of things that's being tested out is a Plogue-based decoder, a bit of a "SPEC for quad decoding folk." I have no clue as to what makes one decoder better than another. If anyone is interested in testing out such a layout, and comparing the output to what's out there already, drop myself or Zeerround a PM.

 

[watzman]

The latest release is the "SQ Plus" provided by oxforddickie on 9-21-2010:

SQ Plus - Improved script available

https://www.quadraphonicquad.com/forums/showthread.php?13482-SQ-Plus-Improved-script-available

But it works only with AA 3.0 AFAIK. I'm not sure if the script could be modified for AA 1.5. But we could ask. You'll also want to get the free VST plugin "Phasebug" at bottom of this link: http://www.betabugsaudio.com/plugs.php

A few day ago I have asked oxforddickie whether he could provide his latest scripts ("SQ350" and "QS360" which are superior to "SQ Plus" / "QS Plus" ones), but so far I did not get any response from him :-(

 

[oxforddickie]

Sorry, there will be no further releases of scripts until they are finished

The complete process is no longer just down to using scripts in Audition. The "Process" is out there with a couple of the long term "trusted" converters at the present.

Once released, there will be a complete "How To" guide along with a list of the software needed to enable a quality decode.

Unfortunatley i can't say when the complete package will be available, there's only so many hours in the day.....

OD