SQ Encoder?

[oxforddickie]

I don't understant what your trying to say with the patents.

Just because Dolby reversed the phase of the rears in a (failed) attempt to not step on the toes of CBS does not make it QS. Qs is totally different. there is fr more to all of the matrix systems than the phase encoding.

I'm now home, so here's something to think about (and i'm doing this for the last time):

SQ:
Lt = Lf + (0.707 Lb -j + 0.707 Rb)Rt = Rf + (-0.707 Lb + 0.707 Rb j)

QS:
Lt = (0.924 Lf + 0.383 Rf) + (0.924 Lb j + 0.383 Rb j)

Rt = (0.383 Lf + 0.924 Rf) + (0.383 Lb -j + 0.924 Rb -j)

Now the two systems are based on totally different ideas, and are totally incompatible. And that's not just because the phase angles used are opposite.
There are better places to discover how the two systems came about, so we'll now jump onto the other system:

Dolby Surround:
Lt = Lf + (0.707 Lb j + 0.707 Rb)

Rt = Rf + (0.707 Lb + 0.707 Rb -j)

You'll notice there is no equation for the center channel as it is a fake, created in the decoder. Typical Dolby!!!!


Now, it is well documented that Dolby stole the idea from SQ by tweaking it to make it incompatible. It didn't work, and ended up with Dolby buying the rights from CBS. As i said previously, just swapping the phase does not make it QS.

I'm now leaving this subject beacause it keeps coming up, it bores me to tears, and nobody actually takes the time to look into it further. They'd rather continue believing the false myths that have grown over the years.


FIN

 

[par4ken]

Apples and oranges again. Dolby Pro Logic II is what you show equations for.

Dolby Surround the original system had only one back channel played back through two band limited speakers. I still have a Fosgate 360 degree Space Matrix decoder sitting in the garage that sounds terrible with the band limiting on the mono surrounds! Dolby Surround encoded the back channel phase shifted +90 degrees in Lt and -90 degrees in the right just like QS. The difference is QS has two back channels differentiated by level differences between L and R. In both systems all encoded signals are in phase (front) or out of phase 180 degrees (back), as in all RM type systems, so it is not what would be referred to as a phase matrix, that is the reported similarity to QS. I do see similarities with SQ as well and acknowledge what you say about (Dolby Pro Logic II) it's clearly an SQ rip off.

The point of the patent drawings is to prove to you three things from the highest source (like the bible). First there is more than one correct way to encode SQ, each is useful for different purposes and each causes different phase issues. Second the fronts always pass through a phase reference network (except for internal encoding) you claimed they did not. Thirdly that my simple encoder and those described by others encodes correctly barring the accuracy of the all pass phase-shift network. You show the correct equations for the basic (four-corner) encoder and that is exactly what my encoder does.

An interesting side note is that the forward oriented encoder can also be used to convert stereo to an SQ encoded signal which can then be processed by an SQ decoder for stereo enhancement. All of this information is in the original patents and I have copies of most of them.

 

[oxforddickie]

I'm afraid your totally wrong, it is the encode equation for DOLBY SURROUND.

THIS is the equation for Dolby PL II:

Lt = Lf + (-j .76 Lr) + (-j .24 Rr)
Rt = Rf + (j .24 Lr) + (j .76 Rr)


Again, i suggest you do your homework

 

[par4ken]

I've done more than my share of homework, and I'm getting tired of homework! If you are correct there must be at least three Dolby matrices then.

The original that I've been referring to as Dolby Surround is Dolby MP. Attached is information right from the horses mouth off the Dolby site. It shows a +j phase shift on the left and -j on the right. Wikipedia shows -j on the left and +j on the right for encoding which conflicts with the information from Dolby. I believe that Dolby would be right. The equations that you show are almost QS with the j terms reversed as well. Where did your Dolby Surround equations come from?

 

[oxforddickie]

I think you need to do more homework. The figures i use are also from an official Dolby document.

Let me explain why your probably confused. The original encoder was a ripped off SQ encoder, hence the equation swap. The obvious difference dolby did was to feed the same mono information to both rear phase circuits. Obviously, once they had to fork out money they reverted to the original phase angles, hence the two different equations. depends on the timeline your looking at.

Either way the results are the same, it's mono so it doesn't matter.

Dolby PL II IS NOT QS. No way is it, do not get confused with the phase angles. As you know Dolby PL II is stereo on the rear, but the way it's encoded is different from both SQ and QS. Yes you can get an effect by playing QS material through a PL II decoder, but a stereo source will also give some effect.

After being burnt once do you really think that they would want problems with sansui? I think not.


So, for the last time, DP II is not QS. The only thing that is... IS QS

 

[par4ken]

OD here are a couple links, the first is correct (I think) regarding Dolby MP (Dolby Surround) http://encyclopedia.thefreedictionary.com/Dolby+Surround the second from Wikipedia must be incorrect, even the coefficients make no sense. http://en.wikipedia.org/wiki/Dolby_Pro_Logic. Also why specify the coefficients so cryptically using square roots? The references for these pages are dead links, I'm at a dead end to finding out more.

And I never said that Dolby PL II was QS, just that it looked similar (regarding the rear channel encode). I think that they stole ideas from both SQ and QS adapting them and possibly using existing equipment as you say in the beginning. Dolby PL II almost looks like SQ for the front and QS for the rear. Also funny that in some of their propaganda they brag about improving on the shortcomings of the 70's Quadraphonic systems. Total BS!

I don't get where you got this from

Dolby Surround:
Lt = Lf + (0.707 Lb j + 0.707 Rb)
Rt = Rf + (0.707 Lb + 0.707 Rb -j)

If this is used for a mono surround encode, then Lb=Rb it will produce a +45 and -45 degree phase shift, 90 degree difference (Lt-Rt), how is that compatible with Dolby or anything else?

 

[oxforddickie]

I don't get where you got this from

Dolby Surround:
Lt = Lf + (0.707 Lb j + 0.707 Rb)
Rt = Rf + (0.707 Lb + 0.707 Rb -j)

If this is used for a mono surround encode, then Lb=Rb it will produce a +45 and -45 degree phase shift, 90 degree difference (Lt-Rt), how is that compatible with Dolby or anything else?[/QUOTE]

I got it from Dolby.

And your maths are wrong, not sure where your getting this 45 degrees from. In maths, j = 90 degrees.

 

[par4ken]

OD do your math; it helps to look at the phaser diagrams a signal at zero degrees coupled with the same signal at 90 degrees (Lb and Rb being the same) results in one signal out at 45 degrees increased in amplitude by 3dB. I would like to see the information from the Dolby document, can you upload it or post a link please.

 

[oxforddickie]

I think you'd best re-evaluate your knowledge. You obviously know nothing about the subject. I'm bored of this, it's going knowhere. If you know so much, why don't you take over fromme in decoding the matrix's. Start with Matrix H

Oh, and by the way, your awswer that backs me up is in what YOU have posted.

Give me strength!!!!!!!!!!!!

 

[par4ken]

No need to be insulting OD. This discussion has taken us on a tangent away from SQ encoding but has raised a lot of questions about Dolby Surround encoding, I just want to know more. Don't get upset that I question some of the things that you say, I do respect your knowledge, I'm just asking to be pointed in the right direction. I've been posting links and documents to support the things that I've been saying and to aid in the discussion. Why can't you help out too, if you are privy to documents that I obviously don't have won't you please share them?

I think that another negative term is required in those equations to be useful in encoding surround.

 

[oxforddickie]

I was getting tired of continually having what i said being questioned. I did mention that this subject had already been dealt with. It was then when you should have gone searching for it.That is what the search function is for.

But, like most people, they don't, and that just pisses me off.

I used to try and help people, but certain 'people' made me feel that i was wasting my time, as they know more than everyone, so all i do now is give information, and leave it at that.

I'm privy to a lot of informtion that i've researched, but for the same reason i see no reason why i should share it. If you think i'm being selfish, well that's your view. There are people here who aren't worth the time of day, and i no longer see why i should provide any help to them.

The reason i know about Dolby Surround is because earlier this year i worked on an improved decoding process, and it was very successful. Both ELP's 'Pictures at an exhibition' and the DS version of Tomita's 'Planets' were much more enjoyable to listen to. I had planned torelease them, just never got round to it.

Maybe one day after 'H' is cracked.

 

[national-kid]

For reference.(It is encoder for adjustment of the decoder)
Considering the electret condenser microphone input.
http://www.hi-ho.ne.jp/odaka/quad/6ch-encode.pdf
Simplify the above circuit.
http://www.hi-ho.ne.jp/odaka/TRGI/encode.pdf

 

[ingresman]

A while ago I looked into creating a free SQ decoder (it's all on this site) but ,my issue was getting an implementation of j i.e. a consistent 90degree shift which all my research points to a Hilbert transform to do. I think I just failed to get my head around the maths behind it so it kind of died.

Doing the encode in realtime (or very near) would be possible in a quick enough machine (and most will be these days) if the Hilbert thing was done right. But if a "batch" encode was ok I think Audition would be ok to do it as you don't have all the seperation steps going. The reason I say very near is lots of the DSP methods involve loading up a period of time worth of samples to work on, so if for example you needed to start up with 1 second of samples you would always have that delay.

So for an encode I think a hardware device woulkd probabbly be the best way forward.

Getting onto the H decode, I recall (tin-hat on here) that a 60degree shift in front of a QS decode does the job.
If anyone has a legal way to get various degrees of shift please post and I can restart my project. I even looked at VST plugins but couldnt find a free one.

I've got a fair bit if H somewhere in my house inclusing WYWH, wish I could find it.

Finally, you all have probabbly already seen this on Tab's site http://www.4channelsound.com/encode.htm

 

[oxforddickie]

Doing the encode in realtime (or very near) would be possible in a quick enough machine (and most will be these days) if the Hilbert thing was done right. But if a "batch" encode was ok I think Audition would be ok to do it as you don't have all the seperation steps going. The reason I say very near is lots of the DSP methods involve loading up a period of time worth of samples to work on, so if for example you needed to start up with 1 second of samples you would always have that delay.

So for an encode I think a hardware device woulkd probabbly be the best way forward.

Getting onto the H decode, I recall (tin-hat on here) that a 60degree shift in front of a QS decode does the job.
If anyone has a legal way to get various degrees of shift please post and I can restart my project. I even looked at VST plugins but couldnt find a free one.

I've got a fair bit if H somewhere in my house inclusing WYWH, wish I could find it.

As to the SQ encoder, a real-time software encoder is not that hard to do

And as to 'H', i'm afraid this myth about the right just needing a phase shift of 60 degrees comes from an incorrect announcement at the time by the BBC. It requires a little more than that.

The reason the truth wasn't put out was because of a commercial agreement between the BBC and the company who took on the job to modify the Sansui receivers.

 

[surteess]

Here is mine, with Marantz tuner above.

The tuner has an AM stereo pcb fitted (designed & built by myself about 30 years ago - using a Motorola chip based on Harris C-QUAM system).
http://en.wikipedia.org/wiki/AM_stereo

I should have turned them both on so you could "feel" the warm glow !!!

 

[par4ken]

I have a Sony ST-JX220A FM Stereo/AM Stereo Tuner, unfortunately there seems to be no AM Stereo left! I had a Plymouth Voyager with AM stereo and it was great, most of those stations have since "flipped" over to FM. When they used to promote AM stereo in the early 80's nobody had anything to receive it with! As AM stereo became more common (at least in automobiles), AM stereo died much like the Quadraphonic situation!

 

[surteess]

Yes still got my good old Sparkomatic SR-430 in the garage, bought it in Toronto in the mid 80s - nice piece of gear (for it's day).

http://mysite.verizon.net/tekel/amstereo/aftermkt.htm
(second to last one on the page).

 

[Q8]

Here is mine, with Marantz tuner above.

The tuner has an AM stereo pcb fitted (designed & built by myself about 30 years ago - using a Motorola chip based on Harris C-QUAM system).
http://en.wikipedia.org/wiki/AM_stereo

I should have turned them both on so you could "feel" the warm glow !!!

Yeah, an "on" pic would be great! *drool* What are you using the encoder for right now? How often do those come up for sale?

 

[surteess]

The setup that I have for the SQ encoder is -

Turntable ---- CD-4 demodulator ---- SQ encoder ---- Sound Forge (software) ---- Burn CD

Some records can only be bought in CD-4 format, prefer to just use CD + decoder....

 

[Q8]

The setup that I have for the SQ encoder is -

Turntable ---- CD-4 demodulator ---- SQ encoder ---- Sound Forge (software) ---- Burn CD

Some records can only be bought in CD-4 format, prefer to just use CD + decoder....

Okay, I have to ask. Why not just make a DTS CD of the CD-4 records instead of making an SQ encoded CD?

Also, how well does that thing encode? Do the SQ encoded CD's decode well and imitate the channel placement on the CD-4?