Can We Please Improve The SQ and QS Scripts....Or Build A Virtual Tate?

[Old Quad Guy]

Again, I wish to thank all of those who worked so hard perfecting the various scripts. These have become my primary decoding sources for Quadraphonic music. In fact, as I write this I'm working on yet another conversion and plan many more.

In the last 4 months since April 2007 I have done literally over 100+ conversions of my record collection using both Adobe Audition 1.5 / 2.0 and many different scripts. I take pride in my work which takes many, many long hours of work only to discard a whole months worth of work because I've found a small improvement or my work was wrong. It's important not to have too much of an ego and keep a "Scientific" point of view about doing conversions. My question is simple: can we please improve the SQ and QS scripts? This is not meant to be critical of all the great hard work already achieved. Far from it. I'm just wondering if we can improve the scripts or is there only so much separation that can be achieved using Adobe Audition?

Recently, I was able to listen to "Oye Como Va" through a Fosgate TATE II Space and Image Composer. The sound separation was unbelievable and was so much more dramatic than the SQ scripts I've been using. I was also able to listen to Four Tops "Keeper Of The Castle" through the Sansui QSD-1000. This is a favorite title of mine that I've recently converted, so I know how it sounds decoded with the various QS scripts. The separation seemed perfect. No bleed through to the other channels with vocals or other instruments. Let me emphasize that the difference between the scripts and vintage machines is like night and day in many respects. You can really hear the differance.

I'm just wondering if we can improve the scripts or is there only so much separation that can be achieved using Adobe Audition? How much db of separation are we getting from the scripts? And are there other factors that need to be considered in order to get the best possible decoding from SQ and QS sources using a computer? Can we build a virtual Tate?

 

[Lucanu]

Hi, I'm still playing with Audition to improve both SQ and QS decoding but now I'm working with ".ses" files so I can hear how SQ or QS are decoded "live".
Playing with scripts takes time to achieve better results so that's how I'm working right now.
My "testing" session files for QS are playing very well.
I just tested "Keeper" Of The Castle" and compared to Q8 version I can say they're pretty close.
I think a "simple" script or a session file couldn't never be fully compared to a Tate or a QSD-1000 (because of their complexity in extraction of information).
Maybe a improved software extraction can be done if we know exactly how those decoders "mathematically" work, but that will be a hard work to do.

 

[Cai Campbell]

Let me emphasize that the difference between the scripts and vintage machines is like night and day in many respects. You can really hear the differance.

So which do you think sounds better? I hear you praising the vintage decoders but at the same time it sounds like you are praising the scripts. I'm a little confused after reading your post about which you ultimately prefer. I honestly want to know. I've been wanting to do a an in-depth comparison between script-decoding and Tate/QSD performance, but have not been able to get around to it, so I would really like to hear your honest and detailed opinion.

 

[Old Quad Guy]

So which do you think sounds better? I hear you praising the vintage decoders but at the same time it sounds like you are praising the scripts. I'm a little confused after reading your post about which you ultimately prefer. I honestly want to know. I've been wanting to do a an in-depth comparison between script-decoding and Tate/QSD performance, but have not been able to get around to it, so I would really like to hear your honest and detailed opinion.

I thought the "air" of separation with the vintage decoders seemed to cut like a knife through to the center of the room. Whereas the scripts seem to lack the same "air" of separation of movement. Mind you, I'm no expert and fairly new to decoding and I'm trying to understand the vintage equipment/scripts as well as the math behind encoding SQ/QS.

Perhaps my 2 different amp and speaker set-up's are crap compared to what is at the Quad shack and that makes the difference. Also, isn't there some sort of "pro logic" like circuits in the vintage equipment to help with the decoding? And do the scripts have that? I was working on a record last week with one of the SQ scripts and the rear channels would have different volume levels with each song. I used WavePurity to solve the volume issues. Does the scripts produce decoding anomalies? Or are the scripts done, completed no need to worry about a thing? I spend most days working on SQ/QS project's as I have loads of source material, but I'm not sure if my work is correct. I worry about that. Sorry for so many questions, but these issues have been on my mind considering the amount of time I spend with the scripts.

I think the idea of an in-depth comparison between script-decoding and Tate/QSD performance is an excellent one and should be explored.

 

[Cai Campbell]

Perhaps my 2 different amp and speaker set-up's are crap compared to what is at the Quad shack and that makes the difference.

The equipment and room/setup differences are potentially major variables. I'd say that you can't draw any reliable conclusions by playing scripted conversions on one system and comparing (via memory) that outcome to Tate/QSD source performance on a different system at a later time. Another major variable is the equipment used to capture the analog signal to begin with. Ideally, the analog capture equipment (e.g. turntable and phono stage) should be the same in both scenarios (script-converted playback vs. source playback through Tate/QSD).

Also, you're invariably going to lose "air" when you convert the analog signal to digital, process the hell out of it, and then subject it to lossy compression (DTS).

Finally, I would go so far as to say much of what you're hearing is just sweet, sweet analog sound. Even if an SQ/QS script were to match or beat a Tate/QSD in decoding, there is no way you're ever going to retain the warmth and purity of the original analog signal. So in the end, even given the best of circumstances, it's a trade-off (purity of sound vs. soundfield stability).

So, I would concentrate less on psycho-acoustical differences and more on technical decoding differences since those psycho-acoustical differences will always be there.

 

[jdmack]

I have never had the opportunity to listen to a SQ or QS recording through a really good decoder (or any decoder for that matter). All I know is what the Audition 1.5 scripts can do and I'm *very* happy with them - with one exception. If there is a short percussive sound, such as lightly tapping on a high hat or solo glockenspiel, the attack of the sound is significantly lengthened. Does this happen with Sansui and Tate II decoders as well? Is this something that might be improved in the Audition scripts one day?

J. D.

 

[Dylan Berichon]

There are advantages (and disadvantages) to both the scripts and the vintage gear. Each have their pros and cons.

Still, with each release of Audition, the filters are improved. Sooner or later we will probably have at our disposal a SQ and QS script that does out perform the Tate and the QRX models in all respects.

That said, there is only so much that can be done at this point in software.

I have come up with a much improved QS script, but it has some problems. (It has a kind of watery sound, but the separation is amazing!) I'm still working on making it better.

 

[Old Quad Guy]

There are advantages (and disadvantages) to both the scripts and the vintage gear. Each have their pros and cons.

Still, with each release of Audition, the filters are improved. Sooner or later we will probably have at our disposal a SQ and QS script that does out perform the Tate and the QRX models in all respects.

That said, there is only so much that can be done at this point in software.

I have come up with a much improved QS script, but it has some problems. (It has a kind of watery sound, but the separation is amazing!) I'm still working on making it better.

Thanks to everyone for their thoughts and hard work. It's good to know that the scripts are still being worked on. Perhaps we can get a progress report from time to time. Or if anyone finds problems or improvements please report them here.

 

[Cai Campbell]

If there is a short percussive sound, such as lightly tapping on a high hat or solo glockenspiel, the attack of the sound is significantly lengthened. Does this happen with Sansui and Tate II decoders as well?

By all means, no!

 

[ChristopherLees]

Perhaps my 2 different amp and speaker set-up's are crap compared to what is at the Quad shack and that makes the difference.
I think the idea of an in-depth comparison between script-decoding and Tate/QSD performance is an excellent one and should be explored.

Have you tried just comparing the lp played in normal stereo on your hifi with the quad vesrion after it's gone through your computer...quality wise...frequency response etc..

 

[Old Quad Guy]

Have you tried just comparing the lp played in normal stereo on your hifi with the quad vesrion after it's gone through your computer...quality wise...frequency response etc..

No, I haven't thought of just playing the record and comparing that with the conversion. That's a good idea. I try not to play records except once or twice to convert because they're irreplaceable. But I'll try your suggestion.

It would be great if we could do an A / B of scripts, Fosgate Tate and Sansui QSD-1000 using aggressive and non-aggressive mixes.

Does anyone have the answer to these questions:

How much db of separation are we getting from the SQ / QS scripts compared with the Fosgate Tate and Sansui QSD-1000?

How does the "pro logic" like circuits in the vintage equipment work to help with the decoding? Thanks.