Quantcast

Audionics Space & Image Composer / Tate Audio

Help Support QuadraphonicQuad:

Cai Campbell

In Remembrance
Since 2002/2003
Joined
Mar 4, 2002
Messages
2,970
Location
Seattle, WA
Dan, please don't feel like I'm trashing your opinion. I'm just stating a contrasting opinion. You've sung the Model 3a and 4's praises on previous occasions and I've never said anything. This time I felt the need to speak up just so folks know there are different viewpoints. I totally respect your viewpoint and like I said, there is no "right" answer. This could just be another "Pet Sounds" situation where I end up as a minority, but that's just the way it goes sometimes.

Actually, Dan, after our little experiment with your Model 4 I was intrigued and actually went out and bought my own so I could experiment further. I thought maybe yours just wasn't calibrated for my system and I wanted to test it thoroughly for myself since it really is a cool unit! Anyway, I picked one up and spent the time to calibrate it properly and then did some extensive listening with it. It did indeed sound better than when we tested yours.

The Fosgate Audionics Model 4 is a really cool decoder and it does a lot of really interesting things. And you know what? Depending on the material being played it can sound better than a Tate or a QSD-1. Overall, though, I found myself preferring the sound of my Tate and/or QSD-1 over the Model 4.

In the meantime, there had been discussion here concerning the Model 3a vs. the Model 4. Well, being the glutton for punishment that I am, I just had to figure out for myself which one I thought sounded better, so I went and found a Model 3a! Long story short, I prefer the Model 4 over the 3a, so I ended up selling the 3a.

The bottom line is all of these decoders do things differently and depending on the source material and your personal tastes, any of them can sound better than the others. If my system were not already overloaded, I would keep the Model 4 in my system. But as it is I just have too much stuff! Something had to go, and since I found myself not using the Model 4 that much, it got yanked. As a matter of fact, it is sitting right here beside me, with the remote. I keep debating whether or not to sell it. I'm not trying to badmouth it. It IS a very, very cool unit! I like it a lot! But choices had to be made...

So that's that story, and yes, some of the tests we performed that day made the QSD-1 sound pretty lousy. Again, these units are highly dependent upon the source material. We were just pulling random stereo material and testing that day. I believe it was a Heart CD that sounded like crap on the QSD-1.
 

Cai Campbell

In Remembrance
Since 2002/2003
Joined
Mar 4, 2002
Messages
2,970
Location
Seattle, WA
The Quadfather said:
I can't speak for Cai, but one time I loaned Tab a copy of Billy Joel "Turnstiles", to process for DTS. My copy was surprisingly flawless, considering that it had gotten lots of play. Tab processed the unit through a Fosgate 101A. When I listened to the DTS version, I noticed speaker switching artifacts that I hadn't noticed when the "master" was played through my own Composer. It seems like Cai was involved with that project somehow, but I can't remember how. I realize it was a limited test, but that's what I got from it. I like the Composer better.

The Quadfather
Nope, I had nothing to do with that. The only time I've collaborated with Tab is when I've sent him material for conversion (which isn't much of a collaboration, really, considering he does all the work!)
 

The Quadfather

1K Club - QQ Shooting Star
Since 2002/2003
Joined
Mar 7, 2002
Messages
1,569
Location
Dixie
O.K. I often forget who has done various things through the years. I probably have a dozen DTS encoded discs that I couldn't tell you who sent them to save my life. But whoever sent them, Thank y'all.

The Quadfather

Cai Campbell said:
Nope, I had nothing to do with that. The only time I've collaborated with Tab is when I've sent him material for conversion (which isn't much of a collaboration, really, considering he does all the work!)
 

sspsandy

600 Club - QQ All-Star
Joined
Jul 9, 2003
Messages
663
Location
Piscataway, New Jersey
I'm very impressed with the decoding abilities of both the Tate and the Model 4. They are different, but both do a fabulous job with synthesizing surround from stereo. I also find the Model 4 does a fine job with SQ material. The Tate was desiged for SQ, so it is more accurate in decoding it. I don't use the Surround mode on my Tate, or Cinema for that matter, I find that the SQ setting is the best for all decoding and surround synthesis. It was Bob Popham who suggested to me that the SQ mode on the Tate was the best arrangement for creating surround due to how it implements the decoding. I'm sure it all depends on how you like to hear things, but for me the Tate in SQ mode does a fantastic job. The Model 4 is excellent, the use of the center channel is a plus, and is quite smooth. Eventually I plan to set one up for the TV room downstairs.
 

deepsky4565

300 Club - QQ All-Star
Since 2002/2003
Joined
Mar 17, 2002
Messages
336
Location
Spokane, WA
Cai, I'm cool, just wanted to know what you were basing your opinions on. I'm glad you did some of your own research, and I do totally respect your opinions. I am sure the Tate sounds different, but when doing my research I found the sound field more pleasing to me on the 3A. For their price, I still think they are worth trying out, and you can always get most if not all your money back on ebay.

I would like to know your opinions after extended listening on the difference between these units and the QSD-1 as far as the soundfield placement. I really like the QS except I do prefer the vocals being front center verses full center.

I'll have to try stereo through SQ mode on my Tate, I just assumed the surround option would be best. Thanks for the tip, Sandy.
 

proufo

1K Club - QQ Shooting Star
Since 2002/2003
Joined
Apr 9, 2002
Messages
1,502
Location
Bogotown, Colombia
Hello again, Steve.

Based on your experience and knowledge of the Tate inner workings, do you believe it is feasible to replicate a Tate II in the digital domain? Maybe software that would not work in real time but search for the best separation parameters by trial and error at all the relevant segments.
 

Steve Kennedy

Member
Joined
Jan 14, 2004
Messages
29
Location
Portland, OR
sspsandy said:
I'm very impressed with the decoding abilities of both the Tate and the Model 4. They are different, but both do a fabulous job with synthesizing surround from stereo. I also find the Model 4 does a fine job with SQ material. The Tate was designed for SQ, so it is more accurate in decoding it. I don't use the Surround mode on my Tate, or Cinema for that matter, I find that the SQ setting is the best for all decoding and surround synthesis. It was Bob Popham who suggested to me that the SQ mode on the Tate was the best arrangement for creating surround due to how it implements the decoding. I'm sure it all depends on how you like to hear things, but for me the Tate in SQ mode does a fantastic job. The Model 4 is excellent, the use of the center channel is a plus, and is quite smooth. Eventually I plan to set one up for the TV room downstairs.
Bob's right on the money there. As far as the Space & Image Composer goes, this is basically how the Stereo Enhance mode works vs. SQ mode so you can see why SQ is closer to Dolby Surround for Motion Pictures (pardon the formatting, the forum won't allow tabs or tables):

The 3 values following the equal sign correspond to the positional output signal in STEREO / Stereo Enhance / SQ

Input Signal

L Ch only = LF / LB / LF
R Ch only = RF / RB / RF
L+R, in phase = CF / CF / CF
L+R, out of phase = CF Cancel / CB / CB
R + L (Quadrature) = * / LS / LB
L + R (Quadrature) = * / RS / RB

* = intermediate positions (psyco-acoustic room decoding inside of your head)

This may not be totally accurate but it's close and illustrates the major difference between SQ and Stereo Enhance (and a clue regarding Solo Null). The Stereo Enhance mode takes a stereo stage and bends it around in a horseshoe, while SQ maintains the stereo stage where it was originally intended and uses quadrature (90 degree phase shifted signals) to obtain surround information.

Dolby Surround started out as a CF reinforced stereo stage with mono surround information. This format more closely resembles SQ than Stereo Enhance and gets even closer in Dolby Surround 2.0 with its' stereo surround speakers. The positional accuracy may not be exact, but Dolby Surround encoded input material will be better served in the SQ mode because it preserves the stereo stage instead of distorting it and places the surround information towards the rear rather than more forward or to the side of the sound-field. SQ derives CF the old fashioned way... L/R imaging psycho-acoustically as in Stereo.

Using Stereo Enhance mode to decode Dolby surround in effect stretches the soundfield that should be across the front into a horseshoe (placing LF & RF into the respective rear speakers) then folds it forward, basically almost reversing the relative locations of the front and surround information. It's more interesting than Stereo but not what was intended!

This is the way I remember it and I'm sticking to it (unless someone who knows I'm wrong points that fact out... this isn't a factually accurate treatise, just a basic comparison to show the concepts at work).

As always, your milegae may vary...

Steve
 

Steve Kennedy

Member
Joined
Jan 14, 2004
Messages
29
Location
Portland, OR
proufo said:
Hello again, Steve.

Based on your experience and knowledge of the Tate inner workings, do you believe it is feasible to replicate a Tate II in the digital domain? Maybe software that would not work in real time but search for the best separation parameters by trial and error at all the relevant segments.
Absolutely! Go back to page 1 of this topic and you will find I posted my thoughts on this earlier! The only difficulty is finding someone who understands the mathematical mechanics of Quad and Surround Sound while also being a competent DSP programmer!

Steve
 

Cai Campbell

In Remembrance
Since 2002/2003
Joined
Mar 4, 2002
Messages
2,970
Location
Seattle, WA
Dan, I do agree with you on the vocal issue. I prefer front-center vocals but don't mind full-centered vocals (it really depends on the material, though). What DOES bother me is when the vocals are pushed too far to the back. Depending on the material, the QSD-1 can push the vocals to the back. A good example of this is Paul McCartney's "RAM". The QSD-1 pushes the vocals so far back that you feel like you should turn around and listen the other way! However, this is the exception and not the rule.

I have played around with listening to stereo material through the Tate in SQ mode, and in some cases, it can sound better than surround mode! Actually, the same is true for the QSD-1. If the soundfield is pushed to far to the rears in surround mode, I will switch to QS mode and the soundfield will come back forward.

The bottom line is (I think this might become my mantra) it depends on the source material! I don't believe there is a single "best" method. I play around with different decoders in different modes using different source material until I get a combo that I really like. Crazy? You betcha! :mad:@:

deepsky4565 said:
Cai, I'm cool, just wanted to know what you were basing your opinions on. I'm glad you did some of your own research, and I do totally respect your opinions. I am sure the Tate sounds different, but when doing my research I found the sound field more pleasing to me on the 3A. For their price, I still think they are worth trying out, and you can always get most if not all your money back on ebay.

I would like to know your opinions after extended listening on the difference between these units and the QSD-1 as far as the soundfield placement. I really like the QS except I do prefer the vocals being front center verses full center.

I'll have to try stereo through SQ mode on my Tate, I just assumed the surround option would be best. Thanks for the tip, Sandy.
 

john e. bogus

Well-known Member
Joined
Jun 6, 2004
Messages
125
Location
USA
Steve-

I found the signal / position information quite interesting.....this seems to imply that the different surround positions can be synthesized or extracted by nothing more than the proper phase shift networks and some op-amps. Extremely accurate phase shifting with virtually no crosstalk can be obtained from common IC's.....it would seem that the challenge would be to obtain a quiet noise floor and flat frequency response to the high end of the audio range. Obviously, it can't just be THAT easy.....what else am I leaving out and failing to consider? Besides the matrix coefficients being different for SQ and QS, what else is changed to obtain the different modes of operation (I'm speaking in the general case here, not of any specific product)? When you state L+R (in phase) or L+R (out of phase), this is with respect to the input signal, correct? (as in L+R in phase simply being the sum of the input signals, and L+R out of phase being the input signals summed and then phase-inverted) Just want to make sure I'm reading your notation correctly.....

Yours Truly,
john e. bogus
 

grimblgrombl

New member
Joined
Aug 11, 2003
Messages
6
Location
Germany
Hi Steve

I'm a proud owner of two S&IC. They work fine but one of them have a really bad seperation. So I recapped this one but the seperation is unimproved. I think it would be great to readjust it.

Could you give us a short instruction how to readjust the Tate logic (inside the S&IC)?


Tobias (Germany)
 
Last edited:

xpablo

Member
Joined
Nov 28, 2004
Messages
33
Location
Penang
what the enhancement ic reference i mean the technique design, if have a reference for that ic maybe we can try figure out to design back that circuit with new changes.
 

Steve Kennedy

Member
Joined
Jan 14, 2004
Messages
29
Location
Portland, OR
john e. bogus said:
Steve-

I found the signal / position information quite interesting.....this seems to imply that the different surround positions can be synthesized or extracted by nothing more than the proper phase shift networks and some op-amps. Extremely accurate phase shifting with virtually no crosstalk can be obtained from common IC's.....it would seem that the challenge would be to obtain a quiet noise floor and flat frequency response to the high end of the audio range. Obviously, it can't just be THAT easy.....what else am I leaving out and failing to consider? Besides the matrix coefficients being different for SQ and QS, what else is changed to obtain the different modes of operation (I'm speaking in the general case here, not of any specific product)? When you state L+R (in phase) or L+R (out of phase), this is with respect to the input signal, correct? (as in L+R in phase simply being the sum of the input signals, and L+R out of phase being the input signals summed and then phase-inverted) Just want to make sure I'm reading your notation correctly.....

Yours Truly,
john e. bogus
I'm no math or SQ expert, but this these are concepts that I have picked up and experienced while being involved in the Composer development. Actually, in my shorthand the (L+R) term is short for the same signal in both channels of the Stereo input signal (not summed). Just as in Stereo, if the same signal is equal in amplitude and phase in both input channels, this signal is heard psycho-acoustically as a Center image. In the Stereo Enhance and SQ matrix, this L+R (in-phase, same amplitude) results in a Center Front (CF) image.

The other mode... (L+R) out of phase... refers to a signal that is equal in amplitude in both the left and right stereo input channels but one channel is 180 degrees out of phase from the other channel. In a perfectly balanced Stereo system at the prime listening position, this results in a perfect psychoacoustic cancellation and theoretically nothing is heard. This can be heard on headphones and when balanced well, resembles a "hole" on the sound-field in the center of your skull... a "sphere of silence" as it were. However, in both Surround and SQ modes, this is decoded as Center Back (CB) and is theoretically played at full level at the rear of the room.

Once you can visualize this, then back off the phase difference between the input channels to +/-90 degrees (quadrature) and you can see that if the channels are in-phase, they are CF, out of phase (180) they are CB, half-way in between (+/-90 degrees) they are Center Top (CT), which is half-way between CF and CB in the room on the imaginary "Scheiber Sphere" that surrounds the prime listening area. So, by continuously varying the proper phase relationships between 0 and 180 (while keeping the amplitudes the same), you can pan a signal from front to back following the path along the sphere over your head, hence the Center Top label. This sounds REALLY cool by the way! (I wish I could give you a specific track to check out!) By adjusting the pan to avoid or "soften" the CT position encoding during this pan you can go directly from CF to CB through the center of the listening position.

If you had a mono tone generator and a network of adjustable phase shifters generating a stereo input signal to a Composer, you could observe the positional phase effects directly (and visually) on the Composer display.

In SQ, the theory (as I understand it) is that as you adjust the relative phases and amplitudes of different signals between the Left and Right input channels (encoding your multi-track and stereo sources), the resulting signal is placed at points that fall on this sphere. This is what is referred to as positional accuracy. If encoding absolute SQ positions (the cardinal points i.e. LF, RF, CF, CT, CB, LB, RB) these should fall at their proper location on the surface of the sphere, and everything else (intermediate positions and complex ambience) falls in 3D space in reference to these points.

The trick isn't just to position these cardinal points on the sphere, it is also to smoothly present ambient effects that have innumerable amplitude and phase differences simultaneously. To achieve this, the directional enhancement system almost has to step "out of the way" so this fine detail isn't yanked this way or that (which will collapse the 3D sound field) and let the matrix work with your brain to do the imaging. This is one of the reasons why an accurate matrix with no enhancement can sound so awesome at one point in the room. As long as the prime focus point is in the center of your head, a matrix by itself can be a truly 3D experience! This is one of the reasons why we put a Separation control on the Composer, to pull the DES back and make it less aggressive when the situation calls for it.

This might also lead you to see why a DSP-based decoder would have to be very fast to fool the ear in the same way that a simple analog SQ matrix can! There is simply a whole lot of ambience going on in most good recordings (natural or manufactured) and you miss it when it gets distorted or over-simplified.

This is also why it is important not to try and achieve hyper-cancellation with a DES. You wind up with a sound-field that can look like a 3D star with only the cardinal points on the sphere but everything else collapsing to the center or an adjacent cardinal point in the matrix. This is not good for that open ambient quality that makes things sound REAL.

In an SQ encoder that is not spatially accurate, the resulting sound-field might not have cardinal points on the sphere at all, but create something more like a football or a cheese danish! I hope this "visualization" helps to illustrate the concept and not get too confusing... that is what all the SQ math is for!

Steve
 

Steve Kennedy

Member
Joined
Jan 14, 2004
Messages
29
Location
Portland, OR
grimblgrombl said:
Hi Steve
I'm a proud owner of two S&IC. They work fine but one of them have a really bad seperation. So I recapped this one but the seperation is unimproved. I think it would be great to readjust it.

Could you give us a short instruction how to readjust the Tate logic (inside the S&IC)?


Tobias (Germany)
I outlined the basic alignment concepts in message #22 of this topic on page #2. The alignment procedure only gets the unit internally balanced (DC balance of the 3 DC control pairs generated by the Tate detector chip) to the point that it can be adjusted by ear from then on. BTW: If you have a well set-up turntable and adjust the Axial Tilt function properly, you can substitute the CBS SQ test record for an SQ encoder in that procedure.

After DC balancing, adjust the 4 mentioned controls using the SQ test record and watching the Composer display to get things looking right visually. Then using real program material your are familiar with, slightly tweak things for the smoothest and best audio performance.

First, get the input balance control properly adjusted (maximum cancellation of CF in the rear channels. I usually started the audio tweaks by adjusting just the internal LF/RF DC balance trim-pot while listening to the front speakers only, then listening to the rear channels only and adjust the CF/CB trim-pot for best/smoothest cancellation of Center Front (lead singer or instrument) in the rear channels, then topping it off with a LB/RB balance adjustment to see if anymore smoothness could be had.

I was using a few SQ classical recordings and some Stereo records that were mixed extremely well to do the audio alignments back at Audionics. I would go back and forth between SQ and Stereo Enhance modes until I was happy with the performance parameters. The albums that come to mind are:

Pat Travers: Heat in the Street
Pink Floyd: The Wall
Michael Omartian: White Horse
Ronny Montrose: Open Fire
Alan Parsons Project: Tales of Mystery & Imagination:Edgar Allan Poe (original mix)

These albums are also extremely effective demo material in the Stereo Enhance mode (but some cuts even great in SQ mode by accident).

To really get a handle on it, you will also need to get the Composer schematics from my web site so the concept (and test points to measure at) might be better visualized. I don't have a step by step set-up procedure beyond what I wrote here as most of it was by ear through the experience of helping develop the unit.

Bob Popham was the only other person at Audionics trained to do the alignment procedure but this can get you close. The "by ear" part is to make sure that decoding is smooth and cancellation isn't so deep that the smoothness and ambient recovery suffers.

Steve
 

john e. bogus

Well-known Member
Joined
Jun 6, 2004
Messages
125
Location
USA
copied and pasted from another thread:

"Man, I hear you. I've spent many hours trying to find anyone, anywhere, to consider remaking these rare IC's. I did, in fact, find one company that remakes IC's, with their slogan "The End Of Obsolescence" http://www.innovasic.com/

I fired off a couple of emails enquiring the ic's but received a response that they were "not interested at the this time". Mabey if we bombarded them with emails, they might be persuaded.

A better bet would be to have someone like Steve Kennedy, who is extremley experienced in this game, send them an email. I remember Quadbob looking into this company after I posted it, but he never replied the outcome. The search continues!!"

Dan-

This certainly sounds interesting! Probably the best way to approach these people is to ask them how much money they would need to set up a production run of chips. In other words, instead of asking "will you repro the chips, and what would you charge for one?", to ask "what would it take to get you to run off a batch?" If they would quote a price, then it would just be a matter of finding enough people to pool resources to make the production run.....there are over 1000 people registered in this forum....if 1000 people kicked in $100, that would probably put us in the ballpark.....surely every Tate owner would gladly ante up $100 to have a spare set of chips....a printed circuit board is easy to etch, all you need is the pattern, which can be posted, downloaded, and printed....this would make it possible to build a Tate or S&IC clone for less than $200.....not to mention that there would be plenty of leftover chips.....between current Tate owners and those who wanted to build one, it should be easy to find enough people to go in on it. The problem, as Steve notes, is that there could be possible legal issues....but wouldn't any contracts or exclusive agreements have expired by now? It's certainly worth checking into....

Yours Truly,
john e. bogus
 

proufo

1K Club - QQ Shooting Star
Since 2002/2003
Joined
Apr 9, 2002
Messages
1,502
Location
Bogotown, Colombia
john e. bogus said:
...there are over 1000 people registered in this forum....if 1000 people kicked in $100, that would probably put us in the ballpark.
And again, wouldn't it be better if the whole thing is emulated in software and definite dts CDs or DVD-As are produced from the best Lps available?

I'd guess that for less than that a programmer can develop the code for the Tate algorithms, and even allow for human optimation.
 

john e. bogus

Well-known Member
Joined
Jun 6, 2004
Messages
125
Location
USA
There would be advantages and disadvantages to either approach....

The advantages of a computer / software based system:
1) Burning the decoded results onto DVD-A (or DTS CD) would mean that the vinyl album only had to be played once.
2) The resulting disc could be played on modern equipment, making the music portable and universal.
3) The software could potentially be cheap, or even free.
4) The software could be updated and improved.
5) The software could provide decoding and synthesis into 5.1 in addition to regular quad.
6) The software could be user-variable, or made to emulate QSD and / or S&IC type decoding.
7) No large initial group investment required.

Disadvantages of a computer / software based system:
1) No realtime playback and decoding. Some people will want an entirely analog signal path in their system.
2) The cost of the burner and soundcard would exceed that of the components required to build a hardware based decoder, at least at the present time....I personally doubt that prices would drop to the point where that would change.
3) The cost of the burning / authoring software would also have to be considered.
4) The cost of the disc playback hardware would have to be considered for someone who dosen't own a player that supports DVD-A and / or DTS.
5) Unless someone with that kind of programming talent was really into it as a labor of love type of thing, they would want to be paid for their time, effort, and skill. We could all end up at the mercy of whatever they felt like charging. Even if the price was minimal, the software would eventually be hacked, cracked, copied, pirated, bootlegged, etc, thus cheating the programmer out of being fairly paid for their work.....it's hard to write and sell good software without it being stolen.
6) The burner, soundcard, and software would all have to be installed and gotten to work correctly.....this might give a bit of trouble to those who aren't very good with computers.

The advantages of a hardware based system:
1) Realtime decoding.
2) The highest cost component would be the chips....the rest of the parts could be purchased for around $50 or so.
3) The chips would be available for repair of existing equipment.
4) No real skill or specialized knowledge would be required to build the decoder. Step by step instructions could be posted on inserting and soldering the parts into the board, etching the board, etc....a parts list could simply be printed out and mailed along with a check to Digikey or some similar outfit.

Disadvantages of a hardware based system:
1) Organizing a group purchase on the order of $100,000 or so to procure the chips would be quite a task.....and someone would have to be trusted with a good chunk of change belonging to many people at once.
2) Playback only on the system containing the decoder.
3) No decoding or synthesis into 5.1. The design is "frozen" due to the nature of the chips and would be difficult if not impossible to improve further.
4) The mechanical work of putting it into a box and making it look like something that belongs in your audio system would be difficult for some people. Caliberation and tweaking would be required for correct performance.
5) We'd have to find out the specs for the chips and what's inside them. Possible legal issues with the rights to the design, patents, license fees / royalities, etc.

Did I miss anything? Doubtless others will think of additional advantages and disadvantages of both approaches.... Seems to be a tossup to me.....I think BOTH approaches have more than enough merit to be worthwhile.....

Yours Truly,
john e. bogus

PS- Steve, your post regarding the directional / positional notations and the "Scheiber Sphere" was GREAT.....quite a lot to digest in that one before the next round of questions (lots of great ideas for neat audio experiments in that one, too). On behalf of everyone, I THANK YOU once again! You've been more than gracious in making this kind of information available and answering our questions.....hopefully more people will jump into the technical end of this discussion. This is the type of stuff that I came to this forum to learn.....perhaps in the future you can suggest some additional technical references that aren't based on high-level calculus?
 

Steve Kennedy

Member
Joined
Jan 14, 2004
Messages
29
Location
Portland, OR
john e. bogus said:
There would be advantages and disadvantages to either approach.... Cut for brevity (SK)

PS- Steve, your post regarding the directional / positional notations and the "Scheiber Sphere" was GREAT.....quite a lot to digest in that one before the next round of questions (lots of great ideas for neat audio experiments in that one, too). On behalf of everyone, I THANK YOU once again! You've been more than gracious in making this kind of information available and answering our questions.....hopefully more people will jump into the technical end of this discussion. This is the type of stuff that I came to this forum to learn.....perhaps in the future you can suggest some additional technical references that aren't based on high-level calculus?
Perhaps someone could track down Wesley Ruggles Jr. and ask HIM about the designs, the legalities and all this. Maybe HE has a stash of Tate chips nobody knows about?

I would suggest looking through some of the material found in the Audionics Shadow Vector patent. It outlines most of the positioning information much better than I did and has Scheiber Sphere illustrations to go along with it!

I'm glad some of that info swimming around in my head has a chance to get out after so many years. I don't claim it is 100% accurate, but it is the concepts that are important. I was hoping it would foster some thought of experimentation! There are so many different ways you could play with this sort of technology to do different things.

It was in this spirit that I came up with the Solo Null idea during Composer development... not because it was helpful for adjusting the input balance control (which it is), but because you could cancel out the vocals, bass and lead instruments that are normally positioned Center Front and hear all the hidden effects, ambience, overdubs and background musicians that are typically buried in a Stereo mix. This was an unexpected use of the SQ matrix and Tate DES that happens to work extremely well!

The original concept was embodied in a device called the Thomson Vocal Eliminator, which basically did the equivalent of reversing the phase of an output channel of a Stereo stylus on a turntable. All the Center information would naturally cancel when the left & right channels were summed to Mono. I used to experiment with this at a younger age. Of course, what was left was just Mono information but it was pretty cool if you were an interested party or a musician.

By stretching this concept a bit and applying it to the Tate DES, I was able to accomplish the same thing but retain a stereo presentation of what audio remained after the CF cancellation which increases your ability to detect hidden musical information and separate individual parts. Basically, the CF control voltage is locked at maximum cancellation and the LB/RB control voltages are locked at minimum cancellation (refer to SAIC Interface schematic). Just a note... Solo Null was the term chosen for this mode at the last minute. If you look at the SAIC Tate DES schematic, this switch was labeled by its' original name, "Background Recovery".

When you consider what the Tate DES does and how well it can be made to work despite the chip fabrication problems, it is amazing that all this was possible with only 3 analog chips 25 years ago!

Steve
 

tcdriver

Well-known Member
Since 2002/2003
Joined
Aug 2, 2002
Messages
190
Location
Valley of Heart's Delight
Another hardware option is to reproduce the circuit using discrete components rather than making an integrated circuit. It would probably be much less expensive.

tcdriver
 

xpablo

Member
Joined
Nov 28, 2004
Messages
33
Location
Penang
tcdriver said:
Another hardware option is to reproduce the circuit using discrete components rather than making an integrated circuit. It would probably be much less expensive.

tcdriver
yes this option might be can use to reproduced the i.c again although
discrete component now are more advance from the past design.
 
Top