question about 5.1 vs quadraphonic simulation

[olds1959special]

It seems flipping the polarity/phase on the rear speakers in a 5.1 setup is a normal part of the film mixing process but doing so in a quadraphonic system will only end up causing the front and rear speakers to cancel each other out. Can someone explain this to me?

 

[DVDdoug]

It seems flipping the polarity/phase on the rear speakers in a 5.1 setup is a normal part of the film mixing process

I don't think so. But if the sound in the rear is completely different it doesn't matter because the phase relationships are uncorrelated.

BTW - The bass will cancel almost completely and at higher frequencies it gets more randomized because of the shorter wavelengths and reflections. In most stereo recordings the bass is the same in both channels.

If you had passive speakers it's easy to flip the polarity of one speaker. But another fairly-easy experiment is to flip the polarity of one channel in a stereo recording digitally with Audacity.

If you download Audacity, chose "download without Muse Hub" if you don't want anything extra.
- Open a regular stereo file in Audacity.
- Click on the 3-dots to the left of the waveform and select Split Stereo Track. This allows you to edit left & right separately.
- Select/highlight one of the tracks.
- Go to Effect -> Special -> Invert.
- Go to File -> Export to create a new file with one channel inverted.*

It's OK with me if you want speakers behind you but you're not REALLY creating "quad" without 4 separate channels or some other processing. AVRs have various settings for "up-mixing". I like to use a "hall" or "theater" setting for some delayed reverb in the rear and the "feel" of a bigger space.



* Another "fun experiment", if you've never done it, its to export in mono after inverting one channel. Everything that's the same in both channels gets subtracted/canceled and since the main vocals are usually in the center, that creates a "vocal reduction" effect.

 

[kyuu]

It seems flipping the polarity/phase on the rear speakers in a 5.1 setup is a normal part of the film mixing process

Is it? What's your source for this? Why would that be done? As DVDdoug says, the phase relationships between the rears and the rest of the channels should be largely uncorrelated. IUf something is panning between the rear and front channels there could be some correlated sound, but the path lengths are certain to be different (assuming it is setup in a standard way) so the phase won't end up the same when it reaches your ears anyway.

The thing about a quadraphonic setup is that the rears are setup in such a way that you could end up with cancellations between the front and rear channels, if I recall correctly. This is one of the reasons why it fell out of favor and the standard went to 5-or-more channels instead.

 

[olds1959special]

Is it? What's your source for this? Why would that be done?

It’s something I read once. It would be done because the rear speakers are facing the opposite direction. But experimenting with inline polarity reversal adaptors I found that I didn't like it in my setup. It made the rear speakers sound thin.

 

[kyuu]

It’s something I read once. It would be done because the rear speakers are facing the opposite direction. But experimenting with inline polarity reversal adaptors I found that I didn't like it in my setup. It made the rear speakers sound thin.

Sounds like you reversed the phase for one of the rear channels but not the other? Correlated bass information getting canceled out on a pair of speakers where one is wired out of phase resulting in "thin" bass is what one would expect.

 

[olds1959special]

Sounds like you reversed the phase for one of the rear channels but not the other? Correlated bass information getting canceled out on a pair of speakers where one is wired out of phase resulting in "thin" bass is what one would expect.

I reversed the phase on both rear speakers and left the phase alone on the front speakers. Then the rear speakers had thin bass.

 

[kyuu]

I reversed the phase on both rear speakers and left the phase alone on the front speakers. Then the rear speakers had thin bass.

Hm interesting. I wouldn't expect the front and rears to have correlated bass for most content. What content were you using for testing?

 

[DVDdoug]

I reversed the phase on both rear speakers and left the phase alone on the front speakers. Then the rear speakers had thin bass.

Right. Like I tried to say above... When the front speaker puts-out a positive pressure wave and the rear speaker the inverted negative pressure wave (or vice-versa), they will cancel.

The same thing will happen with stereo and it won't change if you move the stereo speakers (or move your listening position) so that one stereo speaker is in front and the other behind you.

Hm interesting. I wouldn't expect the front and rears to have correlated bass for most content. What content were you using for testing?

It's not true-quad. He's running the same signals to the front and rear.

...I've never done any experiments like that with surround sound but I'd expect the bass to be the same in all channels, just like it's usually the same in the left & right.

 

[kyuu]

...I've never done any experiments like that with surround sound but I'd expect the bass to be the same in all channels, just like it's usually the same in the left & right.

Now that I really think about it, yeah you're probably right.

 

[olds1959special]

Hafler circuit - Wikipedia

en.wikipedia.org

Maybe because I need to only face the front for this to work? And turn down the volume of the rear speakers?

 

[Wayne Highwood]

I was going to mention the Hafler approach to you, but it seems you've stumbled onto it already. Yes, done properly, the rears need to be 7-8 db lower than the fronts.

What amp(s) are you using? Not all accommodate the Hafler method. And you'll need some way to attenuate the rear level.

 

[olds1959special]

I was going to mention the Hafler approach to you, but it seems you've stumbled onto it already. Yes, done properly, the rears need to be 7-8 db lower than the fronts.

What amp(s) are you using? Not all accommodate the Hafler method. And you'll need some way to attenuate the rear level.

I’m using powered speakers.

 

[JSmith]

flipping the polarity/phase on the rear speakers

Who so many threads on the same topics?

Flip phase on rca cable?

I want to add rear speakers to my setup. Do I need to make custom RCA cables to flip the phase?

www.audiosciencereview.com

question about 5.1 vs quadraphonic simulation

Quadraphonic simulation with four speakers

Is anyone else doing something like this?

www.audiosciencereview.com

JSmith

 

[olds1959special]

Who so many threads on the same topics?

Flip phase on rca cable?

I want to add rear speakers to my setup. Do I need to make custom RCA cables to flip the phase?

www.audiosciencereview.com

Quadraphonic simulation with four speakers

Is anyone else doing something like this?

www.audiosciencereview.com

JSmith

Sorry. I think I got the answers I needed now!

FWIW, I believe rear speakers in a movie theater are calibrated 3 dB lower, but I’m understanding the polarity would NOT be flipped on these in the sound mix. (Unless the sound mixer is going for the Hafler effect?)

 

[Chrispy]

Sorry. I think I got the answers I needed now!

FWIW, I believe rear speakers in a movie theater are calibrated 3 dB lower, but I’m understanding the polarity would NOT be flipped on these in the sound mix. (Unless the sound mixer is going for the Hafler effect?)

It's more the content levels rather than speaker calibration/setup, but down 3-6 dB I think is fairly typical for surround speakers. Have you played around with ambisonics?

 

[Wayne Highwood]

I’m using powered speakers.

Well, that would mean you have to extract the Hafler signal at line level to then feed the rears, which is a slight project:

Simple Surround Sound Decoder

ESP Project Pages - Surround Sound Decoder. Visit my other pages for even more!

sound-au.com

 

[olds1959special]

Well, that would mean you have to extract the Hafler signal at line level to then feed the rears, which is a slight project:

Simple Surround Sound Decoder

ESP Project Pages - Surround Sound Decoder. Visit my other pages for even more!

sound-au.com

This is not just flipping the phase in the rears and lowering the volume, right? I guess I’ll just play all my speakers at 0 phase to keep things easy, but I like lowering the volume of the rears by 3-6dB. They are not the same speakers front and back.

 

[olds1959special]

Just found this:

https://www.reddit.com/r/audiophile/comments/1gpg0o4/comment/lwqc9vc

 

[olds1959special]

Surround channels - Wikipedia

en.wikipedia.org

So what if I sum the L/R for the rears?

 

[olds1959special]

Right. Like I tried to say above... When the front speaker puts-out a positive pressure wave and the rear speaker the inverted negative pressure wave (or vice-versa), they will cancel.

The same thing will happen with stereo and it won't change if you move the stereo speakers (or move your listening position) so that one stereo speaker is in front and the other behind you.


It's not true-quad. He's running the same signals to the front and rear.

...I've never done any experiments like that with surround sound but I'd expect the bass to be the same in all channels, just like it's usually the same in the left & right.

From:

https://www.diymobileaudio.com/posts/5965401/

"Phase is relative. A single speaker playing is neither in phase, nor out of phase. Swap the wires on a single speaker and the result is exactly the same. As far as your amp is concerned there is absolutely no difference at all. You can switch the wires wherever is easiest for the exact same result. People frequently wire speakers out of phase on purpose in an attempt to get the final result to be as close to in phase as possible. So, swap the speaker wires, listen, swap back, listen, and leave them wherever they sound best. In all likelihood the difference will be very subtle, and you may not be able to hear a difference at all. Listen for bass response, if you have better bass response with one configuration over the other, that is the configuration that is most in phase.

Again, in phase and out of phase are the exact same signal, in one case the speaker cone moves outward with the initial signal, in the other case the speaker moves inward with the initial signal. Both sound exactly the same until you start adding other speakers into the mix, and even then having them out of phase with the fronts may actually be the best sounding situation, especially if you don't have DSP to adjust the timing of each speaker."