The SQ shoot-out - The Involve SQ vinyl edition vs. the Fosgate Tate II

QuadraphonicQuad

Help Support QuadraphonicQuad:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
I would guess that if -12dB bleeding is imperceptible then a little bit of pumping in the rears would be too.

AoQ said in the podcast that he sent the rip to OD. What came of that?
 
Last edited:
NOT GONNA GO THERE AGAIN.....ok, yes I will..why do they record in 96K??? ...magick????

This observation makes no technical sense, are you confusing A-D sampling rate with frequency response? What on earth do you think the audio bandwith of a disc cutting head is?!!!
 
Last edited:
I would guess that -12dB bleeding is imperceptible then a little bit of pumping in the rears would be too.

AoQ said in the podcast that he sent the rip to OD. What came of that?

I just need to get around to finishing off the next part of the shootout, which will put the Tate up against OD's decoding. That should be coming this week. I will post links here when they're ready.
 
Did talk to Ohverture about any chance of squeezing more numbers out of it and he was negative about it. I will review this again shortly. Meanwhile please remember that it has been conclusively proven that on multiple simultaneous sound sources the human ability to distinguish separation diminishes dramatically - its one of the basic principals that matrix surround is based on. In practicality it does not make matrix inferior to discrete from a real human perception point of view - from numbers YES. I understand this is controversial but all psychoacoustic studies have verified this.

Regards

Chucky

I understand your explanations about the studies on perception of separation. And that's fine and dandy when dealing with things like test tones or perfect situations. But, what about when you're dealing with mixes that already have a soundfield with the signal blended to multiple channels for the purposes of placement? What I'm getting at, is there may be elements that are panned subtly that already have low separation when measuring a number, and then the encoding and decoding will minimize that even further. I know your studies and results, and numbers you've achieved have left you with the conclusion you've achieved the best compromise, but I do want to point out again, the front center leakage to the rear has been reported numerous times to be audible. I think we are dealing with something here that has proven to be audible.

It also seams that leakage should be easy to target, since it's proven to be out of phase between the 2 back channels.



I haven't heard what targeting this leakage does to the overall presentation, but I understand you feel you've reached the best compromise. Is it perhaps possible to add a separation knob to the SM that allows one the ability to increase the separation to their liking? This could allow one to make their own personal preference to the compromise between separation and presentation. Plus, the best setting probably varies between different recordings. (I suppose for those that feel increasing the separation makes the sound suck, you could label it the suck knob)


That said, I realize that spending time, effort, and resources into a piece of audio equipment for a long obsolete format with a very limited customer base is asking quite a bit. If nothing more can come of this, the Surround Master as it is, is a very worthwhile piece of equipment, and just having this as it is has been great. That you already put this time and effort into this, with so little return, has been great, and I want to make sure you understand that it is very much appreciated, even if we can be critical about these finer details at times.
 
Not quite :). But by oversampling you can reduce the roll-off of your Low Pass Anti-alias filter on the ADC & reconstruction filter on the DAC output, it also spreads the quantisation noise (inherent in any digitising process) over a much wider bandwidth so reducing its effect in the audio range, and you get more samples per signal, which helps in the DAC reconstruction filter. The magic comes when you decimate from a higher sample rate to a lower one and gain resolution, so more bits!

NOT GONNA GO THERE AGAIN.....ok, yes I will..why do they record in 96K??? ...magick????
 
NOT GONNA GO THERE AGAIN.....ok, yes I will..why do they record in 96K??? ...magick????

Vinyl is frequently cut with frequencies above 15kHz filtered out. This is reduce cutter head wear. Half speed (or two thirds speed) mastering allows higher frequencies to be cut whilst still cutting no higher than 15kHz at the time.
 
Hi Proufo

Yes you would be right except that you will perceive pumping louder than the -12 db and that is the issue- its not a little bit.

Regards

Charlie

I would guess that if -12dB bleeding is imperceptible then a little bit of pumping in the rears would be too.

AoQ said in the podcast that he sent the rip to OD. What came of that?
 
Hey Army (lets be on first name basis)

Actually I do agree with all of your comments and I mentioned that we could squeeze as many dB separation as you want.....even 100 dB. It all comes down to a compromise as to what annoys you more - the pumping or the slight leakage. As I and Ohverture are electrostatic image, focus and clarity nutters we have pushed the separation as far as we could before WE could perceive pumping. I agree that many quad nutters in this forum are probably into the separation game more than we are.

I said at the time we agreed to make an SQ version of the SM that we were not going to do a numbers war for the above reason.

Happy to do a tweak in time to crank up the numbers if anyone is desperate for it (for a small programming fee) but the result is it will pump but not smear like the Tate.

Even more regards

chucky

I understand your explanations about the studies on perception of separation. And that's fine and dandy when dealing with things like test tones or perfect situations. But, what about when you're dealing with mixes that already have a soundfield with the signal blended to multiple channels for the purposes of placement? What I'm getting at, is there may be elements that are panned subtly that already have low separation when measuring a number, and then the encoding and decoding will minimize that even further. I know your studies and results, and numbers you've achieved have left you with the conclusion you've achieved the best compromise, but I do want to point out again, the front center leakage to the rear has been reported numerous times to be audible. I think we are dealing with something here that has proven to be audible.

It also seams that leakage should be easy to target, since it's proven to be out of phase between the 2 back channels.



I haven't heard what targeting this leakage does to the overall presentation, but I understand you feel you've reached the best compromise. Is it perhaps possible to add a separation knob to the SM that allows one the ability to increase the separation to their liking? This could allow one to make their own personal preference to the compromise between separation and presentation. Plus, the best setting probably varies between different recordings. (I suppose for those that feel increasing the separation makes the sound suck, you could label it the suck knob)


That said, I realize that spending time, effort, and resources into a piece of audio equipment for a long obsolete format with a very limited customer base is asking quite a bit. If nothing more can come of this, the Surround Master as it is, is a very worthwhile piece of equipment, and just having this as it is has been great. That you already put this time and effort into this, with so little return, has been great, and I want to make sure you understand that it is very much appreciated, even if we can be critical about these finer details at times.
 
Hi Proufo

Yes you would be right except that you will perceive pumping louder than the -12 db and that is the issue- its not a little bit.

Regards

Charlie
AoQs Tate samples showed very low levels of bleeding (pumping, yes).

Did any of the testers specifically recognized objectionable pumping from the Tate?
 
I have been following this thread since its inception and every time I read it my head spins a little.

My first and foremost question would be a simple one: Did the brainiacs that devised SQ back in the early 70's ever get it to work properly. After hearing the discrete master tape from which the SQ LPs evolved, wasn't there even a hint that what the tapes contained and what the SQ matrix discs decoded, which I'm sure was via a state of the art system at the time, had little correlation between each other. I suspect this is why a LOT of artists shied away from committing to the matrix systems and stuck with stereo.

CD~4 was a step in the right direction....but was it really?


I was SO turned off to ALL the Vinyl surround schemes that I immediately switched to QUAD open reel but of course we all know that the 1000's of SQ/QS/CD~4 titles were never subsequently released on QUAD OR.

And my only question today would be if one had a step up phono transformers for their turntables then one would have to go from the turntable to the step up transformer and then to the Involve Surround Master prior to connection to the pre pro and as a purist, way too many wires for me.....and if you're not using higher end cables then there would be 'somewhat' of a signal degradation, IMO, from turntable to pre/pro.

I am in NO way knocking those who love their matrix vinyl but all I can say (and I know this has been shared by many others) is when these old QUAD analogue masters are transferred to QUAD SACD, DVD~A and/or BD~A, the anomalies that plagued the Vinyl matrix replications have been lifted with no artifacts whatsoever.

Hope I didn't offend anybody by putting my two cents worth in, but even though I have a very expensive turntable/arm/cartridge and separate vacuum pump literally sitting in mothballs, I could never go back.
 
all that misses the point of it for me, there are some excellent CBS Quads only ever in SQ LP form (including Billy Joel "Turnstiles", Loggins & Messina "Native Sons", Johnnie Taylor "Eargasm", BT Express "Energy To Burn", Janis Ian "Between The Lines", O'Jays "Survival" & "Family Reunion", Harold Melvin "Wake Up Everybody", Isley Brothers "Harvest For The World", MFSB "Philadelphia Freedom" with hits like "Disco Lady", "At Seventeen", "New York State Of Mind", "I Love Music", "Don't Leave Me This Way" etc)
and i would like to try and squeeze every last drop out of these SQ discs as its all we've got unless/until the discrete 4-ch masters ever make it out on SACD or something.

obviously keep it listenable, keep it musical, keep pumping at bay.. but when you take the resulting decoded rears of an SQ LP once its been processed thru the Surround Master and sum them to mono, post-decode you get no artefacts, just lots of lovely isolated rear channel elements.. so the artefacts must occur when that process happens on the fly I guess?
 
AoQs Tate samples showed very low levels of bleeding (pumping, yes).

Did any of the testers specifically recognized objectionable pumping from the Tate?


I did. Specifically "Sweet Emotion". Am not sure if my results were included in the final tally, as I turned mine in late April, which may have been considered late at that time. I tried to tally the results sifting through the 2 1/2 hour quadquast, and don't think mine were, as I preferred the Surround Master for Sweet Emotion for this reason.

Going through the quadcast and getting the results of what A and B were, for my listening results, I preferred the Surround Master half the time, and the Tate the other half. Per the listening test requests, I purposely didn't isolate channels when auditioning the two matrix decoded tracks.
 
I did. Specifically "Sweet Emotion". Am not sure if my results were included in the final tally, as I turned mine in late April, which may have been considered late at that time. I tried to tally the results sifting through the 2 1/2 hour quadquast, and don't think mine were, as I preferred the Surround Master for Sweet Emotion for this reason.

Going through the quadcast and getting the results of what A and B were, for my listening results, I preferred the Surround Master half the time, and the Tate the other half. Per the listening test requests, I purposely didn't isolate channels when auditioning the two matrix decoded tracks.
Many thanks Wunlow.
 
I was worried about one of the emails getting lost in the shuffle. Between the time I dragged my heels, converting over from yahoo to gmail (due to the constant security nags I was getting every time I checked my email), and always being in a rush, I overlooked that email when I did a search to find any I may have forgotten to save over in the folder for that. I found it, and added it to the tally.

Here's the final count.

Billy Joel - Travelin' Prayer - Involve - 1 Tate - 3
Aerosmith - Sweet Emotion - Involve - 1 Tate - 3
Larry Coryell - Funky Waltz - Involve - 1 Tate - 3
E. Power Biggs - Involve - 2 Tate - 3


I still need to finish up some disc authoring for the Tate vs. software decoding portion of the shoot out, and am a bit tied up the rest of this week and weekend, I should be able to get that ready next week.
 
This is my take on the Tate II and SM units. But first, I would like to say this about Columbia/cbs/Sony. They seem to sabotage their own innovations. When sacd was being promoted it was touted for its compatibility with cd. It could have a stereo sacd layer and a multi-channel sacd layer along with the fully compatible cd layer. So what does Columbia/cbs/Sony do when they introduce the sacd, a single layer stereo sacd without a cd layer with a higher price over cd. They should have at a minimum had a cd layer. When you tout the sacd's compatibility why would you continue to have a dual inventory? They did exactly the same thing with SQ quad. Massive promotion of its compatibility with regular stereo. They raised the price and continued to promote the stereo only versions with a dual inventory. Also, they should not have introduced quad with ineffective decoders. At a minimum they should have introduced it when full logic wave matching decoders became available. Rant over.

First, This was strictly a listening test, I didn't use the shoot-out files, no vector scope or Sound Forge (looks fantastic and expensive with a LONG learning curve) that ArmyofQuad uses to analyze material. My phono amp is connected directly to the SM and its output goes into the multi-channel inputs of my Pre-processor. Connection and disconnection of cables to change sources on my fairly complex system was just not going to happen. Luckily, I came upon a Lyn Anderson(Rose Garden) SQ lp that also included a dvd-audio (24/96) multi-channel version of the album decoded with the Fosgate 101A Tate ll unit. So this was my opportunity to compare a Tate ll decoded album to my SM.

Let me say one has to listen pretty hard for any anomalies that in casual listening may not be noticed. On the Tate version I did notice some pumping, although minimal. On one occasion it momentarily pulled the center front (vocal) to the rear when the unit was steering information to the rear channels. Not at all a bad listening experience. I noticed on one cut that the guitar didn't quite make it to the rear right but was clearly rear right on the SM. To me the SM rear was more stable, absolutely no pumping. The front center vocal bleed on the SM amounted to more ambience to the rear, not at all unpleasant, but different than the Tate. Overall, I preferred the SM in this listening test.

My next tests were to compare the SM with discreet sacd releases, in this case all from Audio Fidelity. The albums were Loggins and Messina (self titled), Loggins and Messina (Full Sail), Jeff Beck Group (self titled), Blood, Sweat & Tears(self titled). My first test was to make sure they were the same mixes. All were the same mixes. I did eliminate the Full Sail sacd from the test because the rear channels were reversed compared to the lp version. I also assumed that the Audio Fidelity albums (like most audiophile releases) were made from the original master mixes, more on this later.

When you think about this, comparing 45 year old SQ lp's pressed from safety copies of the master mix to what amounts to the master tapes isn't exactly a fair test, but read on. First, I made sure speaker levels were matched between the lp and sacd versions. My copies of these albums are mint and quiet and have been in my collection since new (early 70's).

These are just listening tests, not blind or double blind, just switching from one source to the other. So not super comprehensive, but I did spend several hours listening and switching. I would listen to a portion of a cut then listen to the same portion on the other source, switching several times. First test, Loggins and Messina (self titled). This lp acquits itself very well compared to the sacd. In fact, several times I lost track of which source I had selected and had to look and see if it was the lp or sacd. I could not detect (or it was so low) any center front vocal bleeding compared to the sacd. These mixes were so close that I wonder if the Audio Fidelity was actually made from the same safety copy mixes as the lp and not the master mixes. The sacd does perform better, but unless you can switch between them it's really hard to tell. The Jeff Beck was essentially the same with a bit more vocal bleeding. But even with that I lost track and had to look to see which was playing. High praise for these lp's and the SM.

The big disappointment was the Blood, Sweat & Tears(self titled) lp. This was not a prime copy, but acceptable ( I have another copy coming hoping it's better and I will retest). The front vocal bleeding was very noticeable with a tilt to the right front. The channel separation was not nearly as distinct as the sacd. In this case the sacd was incomparably better in every way. So, is this an encoding and/or pressing anomaly. Or is it the SM, I'm guessing the former. The question of center front vocal bleeding seems to be a function of the source material from my experience.

Other observations. I have noticed the front vocal bleeding with a tilt to the right on other lp's. One example: I have the Santana Abraxas lp with some vocals on side 2 with the vocal front right. When compared to the dts quad version the vocal was center front. Compared to another copy of the lp the vocal was more centered with some bleeding. So there can be pressing anomalies and/or encoding problems. Not all albums are cut from the same safety copy of the mix. However, these anomalies aside, I couldn't be more happy with the SM. All in all it is a remarkable achievement.

One more thing, I enjoy the shoot-outs and testing and hope these kind of comparisons continue. But I have one sincere recommendation, set that all aside occasionally and just take the time to sit down and listen to the music and enjoy.
 
Great post riknik, thanks for posting your findings.

And I agree with you about Columbia/Sony. We can wonder forever if they had decided to go all single inventory back in the '70s with any album that had a quad mix, much like Ovation did. And also the SACD should have been a slam dunk, but in retrospect I would imagine they did not have the manufacturing capacity to release only SACD's for those titles with both mixes. And did they HAVE to dump their own format for.............DualDisc? :yikes

Alas, it's all water under the bridge now. But thanks for posting your results. It's interesting to hear how close the SM decodes really are to the discrete AF SACDs.
 
Really great post Riknick!

You are correct re the image issues on Blood sweat and Tears. As mentioned previously even a small 1 db mismatch on level in SQ can make the positioning go whoops!

Its a crap format

Regards

Chucky
 
Back
Top