Surround Master Decode - SQ and QS Encodes of Same Album

[jaybird100]

I'm not sure what the point of such an experiment would be. SQ is at its absolute best when the recording is mixed especially for SQ. The SQ function was added to Involve as an afterthought because many people here requested it. Although it is very good I'm sure that still it could be improved. For one thing using the SM center front to centre back separation should be nearly equal on both but is considerably lower on the SQ decode. My S&IC has about 10 dB better separation (Cf-Cb) than the SM. The Involve decoder was developed based on QS so will naturally function better in that mode. So any such testing would have to favour QS! It would prove only that the SM does a better job on QS, not that QS is inherently better than SQ!

According to what Chucky had once posted, the SQ decoding of the SM is genuine SQ. Having played many SQ records through the SM, I can't disagree. I played the test record that came with my S&IC through the SM, and the results were spot-on

 

[par4ken]

According to what Chucky had once posted, the SQ decoding of the SM is genuine SQ. Having played many SQ records through the SM, I can't disagree. I played the test record that came with my S&IC through the SM, and the results were spot-on

I never said that it wasn't genuine SQ. It works much better than any vintage decoder that I've tried but not better than the S&IC! Less separation centre front to center back, that was pointed out by fredblue, I had to check it for myself (difference of 8 dB not 10 as I previously stated). My point is that a direct comparison of SQ to QS might not be totally valid as the Involve is better at decoding QS which it does even better than the QSD-1.

 

[chucky3042]

I never said that it wasn't genuine SQ. It works much better than any vintage decoder that I've tried but not better than the S&IC! Less separation centre front to center back, that was pointed out by fredblue, I had to check it for myself (difference of 8 dB not 10 as I previously stated). My point is that a direct comparison of SQ to QS might not be totally valid as the Involve is better at decoding QS which it does even better than the QSD-1.

The assessment of "better" is not just a numbers chase. SQ is a wonderful method of hiding surround, it's really hard to extract without sounding mechanical. In various stages of our development of the SQ decoder we could have produced any separation numbers you want.
The issue was relating to image, smearing, pumping etc. Sometimes things sound better winding things back a bit. Also ignoring golden eared bullshit 12 dB is in fact the maximum percievable separation (we tested bunches of people)
Gimmie a while I will dig up the SQ separation results

 

[chucky3042]

I had a comparison listen to the Tate at Rustyandi's man cave comparing it to the SM and I could pick the difference instantly and it was the Tate has a fuzzy image- please note that Rustyandi did not notice the issue (we are all sensitive to different things).

Below are our separation numbers, I have included the "vinyl" addition that allowed for L/R frequency response shifts on some cartridges as compared to a proper recording done to digital media.

 

[par4ken]

I had a comparison listen to the Tate at Rustyandi's man cave comparing it to the SM and I could pick the difference instantly and it was the Tate has a fuzzy image- please note that Rustyandi did not notice the issue (we are all sensitive to different things).

Below are our separation numbers, I have included the "vinyl" addition that allowed for L/R frequency response shifts on some cartridges as compared to a proper recording done to digital media.

Which Tate? I use the S&IC. I actually now have three, two of which are fully functioning. One has the original (Tate1) National chips and the other (Tate II) Exar chips. Neither unit produces fuzzy images, just images that sound fully discrete. Discrete quad only has the edge when all four channels are equally busy simultaneously, no matrix system can replicate that! There is not much difference on my system(s) listening to Mahavishnu Orchestra "Birds Of Fire" SQ decoded by S&IC and the Sony SACD, both sound equally fantastic.

I'm not knocking Involve SQ other than to say that it is not quite as good as a Tate, at least not as good as the S&IC. The Involve evaluation module does blow away most vintage SQ decoders, however. The fixed blend of the Sony SQD-2010 pulles the back speakers closer together (I've always hated SQ blend). Comparing with Involve that blend is painfully audible. I have both units mounted in the same box and so can compare at the push of a button.

The Tate (S&IC) for me provides noticeably better results on SQ than the Involve does however. Involve has what I refer to as slight audible upper midrange glare to the sound. The S&IC is sonically neutral (what goes in comes out). That difference in tonal quality alone biases me toward the Composer. I feel that the overall imaging quality of the Composer is noticeably better as well.

I had never noticed that the Involve SQ had less centre front to centre back separation than regular Involve until fredblue mentioned it and I confirmed it also re-confirming Sonics previous findings. I do agree that separation numbers don't always tell the whole story. The Composer has a Separation control that can be dialed down if necessary. I never (or rarely) used it with the Exar chipped decoder unless I wanted to hear mono from all the speakers, but it helps the original Tate (National) at times when the decoder starts producing audible artifacts.

Back to the gist of this thread comparing QS to SQ. My point is that the Involve is set up to work better (i.e. higher separation) in the Involve mode than in the SQ mode so the test might not be valid. I feel that there a bias here against SQ which I strongly disagree with, perhaps a debate in another thread would be warranted, although that might violate site rules about pitting one system against another. The way the Involve SQ was set up involved some compromises, and I understand that but I'm sure that it could be further improved without sacrificing as much separation.

 

[chucky3042]

Which Tate? I use the S&IC. I actually now have three, two of which are fully functioning. One has the original (Tate1) National chips and the other (Tate II) Exar chips. Neither unit produces fuzzy images, just images that sound fully discrete. Discrete quad only has the edge when all four channels are equally busy simultaneously, no matrix system can replicate that! There is not much difference on my system(s) listening to Mahavishnu Orchestra "Birds Of Fire" SQ decoded by S&IC and the Sony SACD, both sound equally fantastic.

I'm not knocking Involve SQ other than to say that it is not quite as good as a Tate, at least not as good as the S&IC. The Involve evaluation module does blow away most vintage SQ decoders, however. The fixed blend of the Sony SQD-2010 pulles the back speakers closer together (I've always hated SQ blend). Comparing with Involve that blend is painfully audible. I have both units mounted in the same box and so can compare at the push of a button.

The Tate (S&IC) for me provides noticeably better results on SQ than the Involve does however. Involve has what I refer to as slight audible upper midrange glare to the sound. The S&IC is sonically neutral (what goes in comes out). That difference in tonal quality alone biases me toward the Composer. I feel that the overall imaging quality of the Composer is noticeably better as well.

I had never noticed that the Involve SQ had less centre front to centre back separation than regular Involve until fredblue mentioned it and I confirmed it also re-confirming Sonics previous findings. I do agree that separation numbers don't always tell the whole story. The Composer has a Separation control that can be dialed down if necessary. I never (or rarely) used it with the Exar chipped decoder unless I wanted to hear mono from all the speakers, but it helps the original Tate (National) at times when the decoder starts producing audible artifacts.

Back to the gist of this thread comparing QS to SQ. My point is that the Involve is set up to work better (i.e. higher separation) in the Involve mode than in the SQ mode so the test might not be valid. I feel that there a bias here against SQ which I strongly disagree with, perhaps a debate in another thread would be warranted, although that might violate site rules about pitting one system against another. The way the Involve SQ was set up involved some compromises, and I understand that but I'm sure that it could be further improved without sacrificing as much separation.

I will be the first to agree that I am not a fan of SQ and note that we are still recovering from the mental scars of decoding that bastard. I really cannot remember which Tate RustyandI did the demo with......perhaps RustyandI could remind me!

Can I ask what speakers are you running?

We all have different perceptions of what constitutes the perfect sound/ image etc and it comes often down to what we have lived with since being a kid. For example most people think good bass is a thumpy boomy woolly sound and are disappointed when they here flat clean bass (eg BEATS headphones). I have been an electrostatic fella since the age of 15 (with Stax headphones) and the biggest super power of electrostatics is a precise image. I may be picking up on factors that really do not effect you. Please note for distortion perception I defer to Dave the Bitch as he is super sensitive to that, I am not.

Oh I am biased but honestly state that it was very apparent to me.

 

[chucky3042]

I will be the first to agree that I am not a fan of SQ and note that we are still recovering from the mental scars of decoding that bastard. I really cannot remember which Tate RustyandI did the demo with......perhaps RustyandI could remind me!

Can I ask what speakers are you running?

We all have different perceptions of what constitutes the perfect sound/ image etc and it comes often down to what we have lived with since being a kid. For example most people think good bass is a thumpy boomy woolly sound and are disappointed when they here flat clean bass (eg BEATS headphones). I have been an electrostatic fella since the age of 15 (with Stax headphones) and the biggest super power of electrostatics is a precise image. I may be picking up on factors that really do not effect you. Please note for distortion perception I defer to Dave the Bitch as he is super sensitive to that, I am not.

Oh I am biased but honestly state that it was very apparent to me.

Also note that 100% of the development on the SM was done with the aid of electrostatic speakers

 

[par4ken]

I will be the first to agree that I am not a fan of SQ and note that we are still recovering from the mental scars of decoding that bastard. I really cannot remember which Tate RustyandI did the demo with......perhaps RustyandI could remind me!

Can I ask what speakers are you running?

We all have different perceptions of what constitutes the perfect sound/ image etc and it comes often down to what we have lived with since being a kid. For example most people think good bass is a thumpy boomy woolly sound and are disappointed when they here flat clean bass (eg BEATS headphones). I have been an electrostatic fella since the age of 15 (with Stax headphones) and the biggest super power of electrostatics is a precise image. I may be picking up on factors that really do not effect you. Please note for distortion perception I defer to Dave the Bitch as he is super sensitive to that, I am not.

Oh I am biased but honestly state that it was very apparent to me.

All my speakers are homebuilt. In the show us your gear thread I show the speakers in my mancave or dungeon (OK it's just the basement), as well as my "wall of tubes". The speakers in the living room are homebuilt transmission lines I'll be posting about them soon I hope. All speaker sets are matched (four only) and are large. Both systems are bi-amped.

I do share your love of electrostatics, I have a set Magnavox electrostatic phones driven by a homebuilt tube amplifier (circuit designed by Joe Curcio, published in Glass Audio, issue 0).
They are ultra detailed, nuance's and detail that you totally miss with regular headphones knock you on the head!

I always wanted electrostatic speakers but the prices (even for kits) always kept me back, multiply that by two for two sets!

 

[chucky3042]

Which Tate? I use the S&IC. I actually now have three, two of which are fully functioning. One has the original (Tate1) National chips and the other (Tate II) Exar chips. Neither unit produces fuzzy images, just images that sound fully discrete. Discrete quad only has the edge when all four channels are equally busy simultaneously, no matrix system can replicate that! There is not much difference on my system(s) listening to Mahavishnu Orchestra "Birds Of Fire" SQ decoded by S&IC and the Sony SACD, both sound equally fantastic.

I'm not knocking Involve SQ other than to say that it is not quite as good as a Tate, at least not as good as the S&IC. The Involve evaluation module does blow away most vintage SQ decoders, however. The fixed blend of the Sony SQD-2010 pulles the back speakers closer together (I've always hated SQ blend). Comparing with Involve that blend is painfully audible. I have both units mounted in the same box and so can compare at the push of a button.

The Tate (S&IC) for me provides noticeably better results on SQ than the Involve does however. Involve has what I refer to as slight audible upper midrange glare to the sound. The S&IC is sonically neutral (what goes in comes out). That difference in tonal quality alone biases me toward the Composer. I feel that the overall imaging quality of the Composer is noticeably better as well.

I had never noticed that the Involve SQ had less centre front to centre back separation than regular Involve until fredblue mentioned it and I confirmed it also re-confirming Sonics previous findings. I do agree that separation numbers don't always tell the whole story. The Composer has a Separation control that can be dialed down if necessary. I never (or rarely) used it with the Exar chipped decoder unless I wanted to hear mono from all the speakers, but it helps the original Tate (National) at times when the decoder starts producing audible artifacts.

Back to the gist of this thread comparing QS to SQ. My point is that the Involve is set up to work better (i.e. higher separation) in the Involve mode than in the SQ mode so the test might not be valid. I feel that there a bias here against SQ which I strongly disagree with, perhaps a debate in another thread would be warranted, although that might violate site rules about pitting one system against another. The way the Involve SQ was set up involved some compromises, and I understand that but I'm sure that it could be further improved without sacrificing as much separation.

One other interesting point is that I agree that the separation of a matrix decoder cannot match discrete on simultaneous signals on all 4 channels. This is true but what is not mentioned is that human perception of separation of simultaneous sounds deteriorates greatly. Its all about the Haas effect again - our tendency to listen to time slices and pick the first arrival, if you have 4 first arrivals the listener cannot spot direction. So this parameter is vastly overrated.

We did tests on Involve/ QS years ago on simultaneous separation - I have attached them - we are not too shabby!

 

[chucky3042]

All my speakers are homebuilt. In the show us your gear thread I show the speakers in my mancave or dungeon (OK it's just the basement), as well as my "wall of tubes". The speakers in the living room are homebuilt transmission lines I'll be posting about them soon I hope. All speaker sets are matched (four only) and are large. Both systems are bi-amped.

I do share your love of electrostatics, I have a set Magnavox electrostatic phones driven by a homebuilt tube amplifier (circuit designed by Joe Curcio, published in Glass Audio, issue 0).
They are ultra detailed, nuance's and detail that you totally miss with regular headphones knock you on the head!

I always wanted electrostatic speakers but the prices (even for kits) always kept me back, multiply that by two for two sets!

Try
Eraudio.com.au some mighty good kits there!

 

[par4ken]

Try
Eraudio.com.au some mighty good kits there!

Like I said $$$$$$. I would love to have any of those but for a set of four you are over $10,000!

 

[MidiMagic]

To ensure consistency in the mastering process, Thorne took possession of the actual master tapes of the Pink Floyd album Dark Side of the Moon, which caused him considerable paranoia to the extent that he hid the tapes inside his piano, reasoning that it would be unlikely for a burglar to steal such a hefty object or search inside it.

Were these master tapes, or multitrack tapes. This makes a big difference.

- If they were multitracks, then optimum encoding could be done.

- If they were actual masters (4 channel?) then the SQ would have had to be encoded with one of the SQ encoders (4-Corners, Diagonal Split, Front-Oriented, Back-Oriented, or Position).

And I wound NEVER put magnetic tapes in a piano. The frame and the wires are made of easily magnetized (maybe in manufacture) iron and steel.

 

[fizzywiggs41]

The Quadrafile record mentioned by Fizzy had nothing to do with the scenario you described above:
Quadrafile - Wikipedia

Sonic,
In my personal opinion when I first heard the Floyd track "Money" in SQ and compared it with the QS I noticed, that The QS encode was much better.(as it should be).
However when I compared it to the actual EMI SQ album there was also a difference in the SQ quad encode . The EMI pressing (of which I have 2) sounded noticeably improved over the Nippon Columbia SQ pressing used on the Quadraphile album.
The only reason I tried the original SQ pressings ,was that I immediately noticed poor separation on those tracks I was quite familiar with ,having listened to the Floyd Album numerous times.
I also tried the first Tubular Bells track and found the same to be true. The Virgin SQ Album is noticeably better ,and again I have 2 pressings on Virgin.

I think the SQ master wasn't up to par ,at least with the EMI and Virgin vinyl pressings.

My listening equipment was with a Fosgate 101a for SQ.
And a Sansui QSD 2 for QS.
BTW ., I had no issues with either the QS side nor the CD-4 side .(pressed by JVC)

I'm just guessing but Maybe there was a mastering problem with the SQ transfer to vinyl.

Anyway I stand by my findings and would not recommend that album for SQ testing but yes for testing both CD-4 and QS.

UD-4 ....well I did not have the demodulater ,but I did listen to the BMX Matrix.

 

[par4ken]

One other interesting point is that I agree that the separation of a matrix decoder cannot match discrete on simultaneous signals on all 4 channels. This is true but what is not mentioned is that human perception of separation of simultaneous sounds deteriorates greatly. Its all about the Haas effect again - our tendency to listen to time slices and pick the first arrival, if you have 4 first arrivals the listener cannot spot direction. So this parameter is vastly overrated.

We did tests on Involve/ QS years ago on simultaneous separation - I have attached them - we are not too shabby!

I'm not sure just how valid that test is using single tones with a tri-band decoder, I would hope you would be able to get most of them right. No doubt about the Hass effect and our sense of sound direction BUT! Some discrete recordings that have a sound that can't be replicated well via any matrix system. I think of the BTO, Donald Byrd Q8's (now SACD's), and others that have "that Quad sound". Encode any of those by any means and the decode does not match the original while other less intense mixes match very closely.

 

[fizzywiggs41]

Tell me how that is an objective listening test? Is it blind A/B or A/B/X? QS encoding is pretty straight forward but SQ... basic, forward oriented, back ward oriented? QS decoded by vintage Sansui, & maybe Fosgate SQ Tate for SQ? Or both decoded by the Surround Master?

Yes both by the Surround Master (version 2)

 

[MidiMagic]

I had a comparison listen to the Tate at Rustyandi's man cave comparing it to the SM and I could pick the difference instantly and it was the Tate has a fuzzy image- please note that Rustyandi did not notice the issue (we are all sensitive to different things).

Below are our separation numbers, I have included the "vinyl" addition that allowed for L/R frequency response shifts on some cartridges as compared to a proper recording done to digital media.

It took some time to read this, sorting r right and r rear.

That's why I never use r for rear. I use b for back.

And I know what discrete quad is, but what is this "discreet" quad. Does it remove indiscreet content.

 

[par4ken]

It took some time to read this, sorting r right and r rear.

That's why I never use r for rear. I use b for back.

And I know what discrete quad is, but what is this "discreet" quad. Does it remove indiscreet content.

I agree, I'm old school I like to call the back speakers back not rear. Back is more politically correct than rear anyway!

Discreet quad is when you listen to quad in private, with no one else knowing!